feat(cli): explode archive materials in chainloop att add by javirln · Pull Request #3254 · chainloop-dev/chainloop

javirln · 2026-06-30T07:31:14Z

Summary

chainloop attestation add --kind <KIND> --value <archive> now unpacks a .zip, .tar, .tar.gz, or .tgz for the kinds that have a meaningful "bundle of the same kind" archive form and records each contained regular file as its own material of the given kind, instead of requiring one att add invocation per file.

Behavior

Explosion applies only to an allowlist of kinds: SBOM_CYCLONEDX_JSON, SBOM_SPDX_JSON, and SARIF. For one of these, when the value resolves to a supported archive, each contained regular file is added as an individual material of that kind.
Every other kind (e.g. ARTIFACT, EVIDENCE, ZAP_DAST_ZIP) records the archive whole, so a customer can still provide a regular zip as a single material.
Per-entry material names are sequential and 0-indexed: material-0, material-1, … with no --name, or <name>-0, <name>-1, … when --name is provided (used as the prefix, sanitized to DNS-1123). Names already present in the attestation are skipped. The recorded artifact filename preserves the original entry basename.
Annotations passed via --annotation are applied to every extracted material.
The operation is atomic: either all qualifying entries are added or none.
Extraction is guarded against zip bombs and path traversal: --max-extract-entries (default 10000) and --max-extract-size (default 1GiB) are enforced while streaming, and directory, symlink, absolute, and path-traversal entries are skipped or rejected.

For non-explodable kinds, non-archive values, and no --kind, behavior is unchanged. There are no server-side, contract, or gRPC changes.

Assumptions and scope

Explosion is limited to the SBOM/SARIF allowlist above; it is intentionally conservative so a regular zip provided for any other kind is never expanded. New kinds can be added to the allowlist as they gain a bundle form.
Every extracted file is recorded under the same --kind; archives mixing material types are not split by type.
Nested archives are not recursed — an archive contained inside the archive is treated as a regular file.
--policy-input-from-file is ignored on the explode path (a warning is logged); policy inputs and evidence cross-links are not recorded for exploded materials.
Zero-length entries are skipped and produce no material.
Only stdlib-supported formats are handled (.zip, .tar, .tar.gz, .tgz); bzip2/xz are not supported in this iteration.

AI disclosure

This contribution was assisted by Claude Code.

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4 Signed-off-by: Javier Rodriguez <javier@chainloop.dev>

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Strengthen path validation in archive extraction to explicitly reject: - Absolute paths (e.g., /etc/passwd) - Relative path traversal attempts (e.g., ../ or foo/../..) Added direct unit test of safeArchivePath and integration test via tar with traversal entry to verify rejection. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Adds an exported AddMaterialsFromArchive method on the Crafter that walks every entry in a zip/tar/tar.gz archive, stages each entry as an independent material via stageMaterial, and commits the in-memory state in a single stateManager.Write call. If any entry fails, previously staged entries are rolled back so no partial state is ever persisted. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

…ry on archive expand Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Add MaxExtractEntries/MaxExtractSize to AttestationAddOpts and wire them into AttestationAdd with defaults from materials.DefaultArchiveLimits(). Change Run to return ([]*AttestationStatusMaterial, error) and insert an explode branch before the single-material switch: when --kind is set and the value is a non-archive-native archive, delegate to crafter.AddMaterialsFromArchive and return N results; otherwise the single-material path continues unchanged and its result is wrapped in a 1-element slice. Add shouldExplode helper and TestShouldExplode. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Add TestAddMaterialsFromArchiveBehavior to crafter_test.go covering five end-to-end scenarios via AddMaterialsFromArchive: name collision suffixing (scan-json / scan-json-1), name prefix, skipping dirs and symlinks in tar.gz, path-traversal rejection with atomic rollback, and tar.gz happy path with two materials. Fixtures are built programmatically with buildZip/buildTarGz helpers using t.TempDir(); no binary fixtures committed. Also regenerate app/cli/documentation/cli-reference.mdx to include the --max-extract-entries and --max-extract-size flags added in earlier tasks. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

… tests Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

…uards - detectByMagic now returns (ArchiveNone, nil) when os.Open fails so non-file material values (STRING, CONTAINER_IMAGE) never produce a spurious "no such file" error through the shouldExplode path. - safeArchivePath drops the over-broad strings.Contains(name, "..") early-return that wrongly rejected legitimate filenames like foo..bar.json; traversal detection now relies solely on path.Clean against a virtual root (/root/), which is the only reliable check. - Add a Warn log when --policy-input-from-file is supplied with an archive value so users know evidence cross-links are skipped on the explode path. - Name per-entry temp files with the allocated unique material name (matName) instead of filepath.Base(name) to eliminate basename collisions; remove each temp file immediately after staging to keep disk usage bounded. - Extend TestDetectArchive with .tar and .tgz cases; add writeTar helper mirroring writeTarGz without the gzip layer. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

Render multi-material explode output as a single JSON array so --output json stays a parseable document; only the table renderer is emitted per material. Drop the unused size parameter from the archive visit signature, document the intentional zero-length-entry skips and the zip symlink-detection limitation, and correct the --max-extract-size default label to 1GiB. Build the crafter archive test fixture in-process and drop the checked-in binary blob and SDD process artifacts. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

chainloop-platform · 2026-06-30T07:31:23Z

AI Session Analysis

Missing AI Coding Sessions

We detected commits in this PR that were AI-assisted, but the matching Chainloop Trace session(s) could not be found in Chainloop.

Please make sure the AI coding session evidence has been sent by the Chainloop CLI, or add the skip-ai-session label to this PR to bypass this check.

Learn more about Chainloop Trace.

Powered by Chainloop and Chainloop Trace

javirln · 2026-06-30T07:41:25Z

End-to-end examples

Verified by running the locally built CLI against a control plane (dry-run / local state). Server side is unchanged, so this needs only a CLI build.

Explode a zip into one material per file

$ chainloop attestation add --kind SBOM_CYCLONEDX_JSON --value sboms.zip
INF material(s) added to attestation materials=3

sboms.zip containing service-a.cdx.json, service-b.cdx.json, service-c.cdx.json produces three independent materials, each crafted as SBOM_CYCLONEDX_JSON with its own digest:

Name    service-a-cdx-json   Type SBOM_CYCLONEDX_JSON   Digest sha256:304ad206…
Name    service-b-cdx-json   Type SBOM_CYCLONEDX_JSON   Digest sha256:51537faa…
Name    service-c-cdx-json   Type SBOM_CYCLONEDX_JSON   Digest sha256:6440a90c…

Behaviors verified

Command	Result
`att add --kind SBOM_CYCLONEDX_JSON --value sboms.zip`	3 materials, each `SBOM_CYCLONEDX_JSON`
`att add --kind ARTIFACT --value collision.zip` (`scan.json`, `nested/scan.json`, `results.xml`)	names `scan-json`, `scan-json-1`, `results-xml` (basenames sanitized to DNS-1123, collisions suffixed)
`att add --kind ARTIFACT --name release --value bundle.tar.gz`	names `release-app-bin`, `release-config-yaml`, `release-readme-txt` (`--name` used as prefix)
`att add --kind ZAP_DAST_ZIP --value zap_scan.zip`	1 material — archive-native kind recorded whole (allowlist short-circuit)
`att add --kind ARTIFACT --value app.bin` (not an archive)	1 material — unchanged behavior
`att add --kind STRING --value "a literal string"`	1 material — non-file values are not treated as archives
`att add --kind ARTIFACT --value many.zip --max-extract-entries 3` (8 entries)	`ERR adding materials from archive: … archive exceeds the maximum number of entries` (nothing added — atomic)
`att add --kind ARTIFACT --value evil.tar.gz` (entry `../escape.txt`)	`ERR adding materials from archive: … unsafe entry path in archive: "../escape.txt"` (nothing added — atomic)

cubic-dev-ai

7 issues found across 8 files

_{Reply with feedback, questions, or to request a fix.

Re-trigger cubic}

javirln · 2026-06-30T07:48:35Z

Additional examples — across material types

Each archive was built from the repo's own materials/testdata fixtures, so every entry is crafted and validated as the given --kind. Mix of .zip and .tar.gz.

Command	Result	Derived materials
`att add --kind SBOM_CYCLONEDX_JSON --value sboms.zip`	3	`service-a-cdx-json`, `service-b-cdx-json`, `service-c-cdx-json`
`att add --kind SBOM_SPDX_JSON --value spdx.zip`	3	`sbom-spdx-json`, `container-json`, `described-json`
`att add --kind SARIF --value sarif.tar.gz`	2	`codeql-sarif`, `semgrep-sarif`
`att add --kind OPENAPI_SPEC --value openapi.zip`	3	`v2-0-yaml`, `v3-0-json`, `v3-1-yaml` (mixed OpenAPI versions, each validated)
`att add --kind GITLEAKS_JSON --value gitleaks.zip`	2	`clean-json`, `violations-json`
`att add --kind OPENVEX --value openvex.zip`	2	`vex-a-json`, `vex-b-json`
`att add --kind ATTESTATION --value attestation.tar.gz`	2	`dsse-json`, `bundle-json`
`att add --kind EVIDENCE --value evidence.zip`	2	`approval-txt`, `review-json`

Each entry is crafted with the same --kind and validated independently. A single invalid entry fails the whole add atomically:

$ chainloop attestation add --kind SBOM_CYCLONEDX_JSON --value mixed.zip
# mixed.zip = one valid CycloneDX SBOM + one non-SBOM file
ERR adding materials from archive: expanding archive "mixed.zip": processing entry "bad.json": \
    staging entry "bad.json" as material "bad-json": crafting material: invalid cyclonedx sbom file: unexpected material type
# attestation status afterwards: 0 materials (atomic rollback)

Validate --max-extract-size before the uint64->int64 cast so oversized values are rejected instead of wrapping negative. Check the per-entry temp file Close error before staging so a failed flush never produces an incomplete material, and write each entry into its own temp subdirectory under its original basename so the recorded artifact filename keeps the real name rather than the sanitized material key. Reject Windows drive-letter absolute paths in safeArchivePath, and only swallow not-found errors in detectByMagic so permission/I/O errors surface. Treat non-positive extraction limits as use-default. Assert the exact detected archive format and the preserved filename in tests. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

cubic-dev-ai

1 issue found across 7 files (changes from recent commits).

_{Reply with feedback, questions, or to request a fix.

Re-trigger cubic}

A non-file material value whose first path segment is an existing regular file (e.g. a CONTAINER_IMAGE ref like "registry/app:v1") yields ENOTDIR on open rather than not-found. Swallow it alongside fs.ErrNotExist so such values detect as non-archive instead of failing detection; genuine permission/I/O errors still surface. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

…tics filepath.Base treats backslash as a separator only on Windows, so an entry name embedding a literal backslash produced different material names across platforms. Add materials.ArchiveEntryBaseName, which normalizes separators and uses path.Base, and derive the per-entry basename through it so exploded material names are identical on every OS. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

migmartri · 2026-06-30T14:21:28Z

Thanks, let's make the names based on the original zipped material name

Per-entry material names are now sequential rather than derived from the entry basename: material-1, material-2, … with no --name, or <name>-1, <name>-2, … when --name is provided (used as the prefix). The original entry filename is still preserved in the recorded artifact metadata. Replaces NameAllocator.Allocate with AllocateSequential. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

cubic-dev-ai

1 issue found across 4 files (changes from recent commits).

Prompt for AI agents (unresolved issues)


Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="pkg/attestation/crafter/crafter.go">

<violation number="1" location="pkg/attestation/crafter/crafter.go:873">
P2: Archive-expanded material names use sequential numbering instead of deriving from entry filenames, directly contradicting the PR description which states "Per-entry material names derive from the entry filename, sanitized to DNS-1123 with collision suffixes."</violation>
</file>

_{Tip: Review your code locally with the cubic CLI to iterate faster.

Re-trigger cubic}

cubic-dev-ai · 2026-06-30T14:46:47Z

+		// (with archive "/" semantics, OS-independently) and used for the temp
+		// file so the recorded artifact filename preserves the real name.
+		base := materials.ArchiveEntryBaseName(name)
+		matName := allocator.AllocateSequential(namePrefix)


P2: Archive-expanded material names use sequential numbering instead of deriving from entry filenames, directly contradicting the PR description which states "Per-entry material names derive from the entry filename, sanitized to DNS-1123 with collision suffixes."

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At pkg/attestation/crafter/crafter.go, line 873: <comment>Archive-expanded material names use sequential numbering instead of deriving from entry filenames, directly contradicting the PR description which states "Per-entry material names derive from the entry filename, sanitized to DNS-1123 with collision suffixes."</comment> <file context> @@ -865,10 +865,12 @@ func (c *Crafter) AddMaterialsFromArchive( + // file so the recorded artifact filename preserves the real name. base := materials.ArchiveEntryBaseName(name) - matName := allocator.Allocate(namePrefix, base) + matName := allocator.AllocateSequential(namePrefix) // Give each entry its own temp subdirectory (named by the unique material </file context>

Exploded archive materials are now numbered from 0 (material-0, material-1, … or <name>-0, <name>-1, …) instead of 1. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

Avoid the repeated "material" string literal (goconst). Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

javirln · 2026-06-30T14:54:17Z

Manual end-to-end verification

Run against the labs control plane + CAS (devel/compose.labs.yml) with a locally-built CLI, driving the real chainloop attestation add surface.

Fixture evidence.zip: scan.json (7B), nested/report.sarif (11B), summary.txt (2B), empty.txt (0B), dir/ (directory entry).

chainloop attestation init --workflow wf --project proj
chainloop attestation add --kind ARTIFACT --value evidence.zip

Observed results:

Explode (no --name) → 3 materials material-0, material-1, material-2; empty.txt and dir/ skipped (5 entries → 3 materials). The material Value shows the original filenames (scan.json, report.sarif, summary.txt), confirming the recorded artifact name is preserved.
--name foo → foo-0, foo-1, foo-2.
--annotation team=sec → applied to every extracted material.
-o json → a single valid JSON array of the materials (not concatenated top-level objects).
Archive-native kind --kind ZAP_DAST_ZIP → single-material path (errors validating the whole zip as one ZAP report), i.e. not exploded.
Guard --max-extract-entries 1 on the 3-file zip → archive exceeds the maximum number of entries, and attestation status reports 0 materials, confirming atomic rollback.
Collision-skip: adding into an attestation that already contained material-1/2/3 produced material-0, material-4, material-5, skipping the in-use names.

Fixes revive redefines-builtin-id on the archive behavior test. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

migmartri · 2026-07-01T12:03:32Z

  chainloop attestation add --name sigcheck --value sigcheckResult.csv --kind SYSINTERNALS_SIGCHECK \
    --policy-input-from-file ignored_paths=exception.csv:Path`,
 		RunE: func(cmd *cobra.Command, _ []string) error {
+			maxExtractSizeBytes, err := bytefmt.ToBytes(maxExtractSize)


how do you know it's something to extract?

It no longer explodes on detection alone — only when the kind is in the SBOM/SARIF allowlist AND the value sniffs as an archive (e358fb8).

migmartri · 2026-07-01T12:05:08Z

+			}
+			// Guard against the uint64->int64 cast wrapping negative, which would
+			// later surface as a misleading "archive too large" error.
+			if maxExtractSizeBytes > math.MaxInt64 {


this shoudl not apply to all materials just when we know taht we want to untar smth. Customers can provide a regular zip file.

Agreed. As of e358fb8 explosion only applies to the SBOM/SARIF allowlist; a regular zip for any other kind is stored as a single material.

migmartri · 2026-07-01T12:11:49Z

+// n is a zero-based counter that advances across calls and skips names already
+// in use. prefix is sanitized to DNS-1123; an empty or symbol-only prefix yields
+// the base "material" (so entries are named material-0, material-1, …).
+func (a *NameAllocator) AllocateSequential(prefix string) string {


this might exist already

Right — consolidated in e358fb8. materials.SanitizeMaterialName is now the single sanitizer (returns the bare component, empty when nothing usable), and the action layer's sanitizeMaterialNamePart delegates to it with its own "input" fallback instead of duplicating the logic.

Drop the redundant explode bool from shouldExplode (derivable from the returned ArchiveFormat) and branch on the format at the call site. Reuse output.EncodeOutput for the multi-material result instead of a hand-rolled format switch, adding []*AttestationStatusMaterial to the tabulated-data union so the slice renders as a single JSON array. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

migmartri · 2026-07-01T12:12:21Z

+	return nil
+}
+
+func walkTar(p string, gzipped bool, visit func(name string, r io.Reader) error) error {


we don't have tar support?

Tar is supported — walkTar covers .tar, .tar.gz and .tgz (gzip-wrapped). See the note on the zip thread re: reusing existing private readers.

migmartri · 2026-07-01T12:12:31Z

+	return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
+}
+
+func walkZip(p string, visit func(name string, r io.Reader) error) error {


no zip support?

We do — walkZip handles .zip and walkTar handles .tar/.tar.gz/.tgz. There was no shared/exported archive walker to reuse (zap.go, helmchart.go and radamsa.go each inline their own private zip/tar reading), so WalkArchiveEntries is the new shared helper in the materials package. Happy to refactor those existing readers onto it as a follow-up if you'd like.

I mean if we don't have it already somewhere else

Address review feedback: explode only kinds with a meaningful bundle form (SBOM_CYCLONEDX_JSON, SBOM_SPDX_JSON, SARIF) instead of exploding any archive that is not archive-native. Every other kind (ARTIFACT, EVIDENCE, ZAP_DAST_ZIP, …) records the archive whole, so a regular zip provided as a single material is never expanded. Replaces the archive-native denylist with an explodable-kind allowlist. Also consolidate name sanitization: SanitizeMaterialName now returns the bare sanitized component (empty when nothing remains) and callers supply their fallback, so the action layer's sanitizeMaterialNamePart reuses it instead of duplicating the logic. Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: da72e107-14e9-4da1-add0-28004f542628, ef6d3cdb-5a23-445c-b39a-510b659023e4

javirln added 14 commits June 29, 2026 17:45

docs: add archive explode implementation plan

336dd0e

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

feat(cli): detect archive materials for att add explode path

5b658e4

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4 Signed-off-by: Javier Rodriguez <javier@chainloop.dev>

feat(cli): walk archive entries with zip-bomb and traversal guards

38800a6

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

feat(cli): add archive-native allowlist and entry name allocation

c5069f1

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

refactor(crafter): split material crafting from state persistence

e0dfe32

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

fix(crafter): roll back policy evaluations and name temp files by ent…

3a17c4e

…ry on archive expand Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

feat(cli): add archive extraction guard flags and multi-material output

2e1450d

Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

fix(crafter): export ErrUnsafeEntry sentinel and assert it in archive…

e7a581b

… tests Assisted-by: Claude Code Signed-off-by: Javier Rodriguez <javier@chainloop.dev> Chainloop-Trace-Sessions: ef6d3cdb-5a23-445c-b39a-510b659023e4

cubic-dev-ai Bot reviewed Jun 30, 2026

View reviewed changes

Comment thread pkg/attestation/crafter/materials/archive.go Outdated

javirln added 2 commits June 30, 2026 11:45

cubic-dev-ai Bot reviewed Jun 30, 2026

View reviewed changes

javirln added 2 commits June 30, 2026 16:51

javirln marked this pull request as ready for review June 30, 2026 15:19

javirln requested a review from a team July 1, 2026 10:55

migmartri previously approved these changes Jul 1, 2026

View reviewed changes

migmartri reviewed Jul 1, 2026

View reviewed changes

Comment thread app/cli/pkg/action/attestation_add.go Outdated

migmartri reviewed Jul 1, 2026

View reviewed changes

Comment thread pkg/attestation/crafter/materials/archive.go Outdated

javirln marked this pull request as draft July 1, 2026 12:11

migmartri reviewed Jul 1, 2026

View reviewed changes

javirln dismissed migmartri’s stale review via a25c290 July 1, 2026 12:11

migmartri reviewed Jul 1, 2026

View reviewed changes

Uh oh!

Conversation

javirln commented Jun 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Behavior

Assumptions and scope

AI disclosure

Uh oh!

chainloop-platform Bot commented Jun 30, 2026

AI Session Analysis

Missing AI Coding Sessions

Uh oh!

javirln commented Jun 30, 2026

End-to-end examples

Explode a zip into one material per file

Behaviors verified

Uh oh!

cubic-dev-ai Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

javirln commented Jun 30, 2026

Additional examples — across material types

Uh oh!

cubic-dev-ai Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

migmartri commented Jun 30, 2026

Uh oh!

cubic-dev-ai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

cubic-dev-ai Bot Jun 30, 2026

Choose a reason for hiding this comment

Uh oh!

javirln commented Jun 30, 2026

Manual end-to-end verification

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

javirln commented Jun 30, 2026 •

edited

Loading

cubic-dev-ai Bot left a comment •

edited

Loading

cubic-dev-ai Bot left a comment •

edited

Loading