chore(deps): update pulumi/agent-skills digest to 641c2fe

[renovate]

This PR contains the following updates:

Package	Update	Change
pulumi/agent-skills	digest	d083ad2 → 641c2fe

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • Between 12:00 AM and 03:59 AM, only on Monday (* 0-3 * * 1)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.

[toolhive-release-app]

🛡️ Skill Security Scan Results

✅ package-usage

• Status: Passed
• Findings: 1
• Allowed (not blocking): 1
  • MANIFEST_MISSING_LICENSE (Allowed: pulumi/agent-skills is licensed Apache-2.0 at the repository root; upstream does not embed an SPDX license identifier in per-skill SKILL.md frontmatter.)

✅ provider-upgrade

• Status: Passed
• Findings: 12
• Allowed (not blocking): 5
  • MANIFEST_MISSING_LICENSE (Allowed: pulumi/agent-skills is licensed Apache-2.0 at the repository root; upstream does not embed an SPDX license identifier in per-skill SKILL.md frontmatter.)
  • ATR_2026_00050 (Allowed: FP: cisco-ai-skill-scanner matched the phrase 'Same error persists' in a bulleted troubleshooting list (SKILL.md, Preview Progress section) describing how to interpret repeated pulumi preview errors. Plain documentation prose, no executable content.)
  • ATR_2026_00066 (Allowed: FP: cisco-ai-skill-scanner matched ${VERSION}/schema-tools-${VERSION}-${OS}-${ARCH} in the documented schema-tools download example (references/diagnostic-toolbox.md); standard shell variable substitution for a GitHub release URL, no injection risk.)
  • COMPOUND_EXTRACT_EXECUTE (Allowed: The skill documents standard npm install / pip install / go get upgrade workflows which involve extracting packaged archives and running their lifecycle scripts. The scanner itself notes 'found in documentation — may be instructional'.)
  • ATR_2026_00064 (Allowed: cisco-ai-skill-scanner matched 'chmod +x' in the documented schema-tools install script (references/diagnostic-toolbox.md:48) (rule renumbered from ATR_2026_00040 in this scanner version). This allowlisting addresses only the privilege-escalation pattern match: making a just-downloaded binary executable is not itself a privilege-escalation technique. It does not establish trust in the downloaded schema-tools release binary's provenance/integrity - that is a separate supply-chain question (no checksum verification is documented in this step) and is not resolved by this entry.)

✅ pulumi-automation-api

• Status: Passed
• Findings: 0

✅ pulumi-best-practices

• Status: Passed
• Findings: 3
• Allowed (not blocking): 1
  • MANIFEST_MISSING_LICENSE (Allowed: pulumi/agent-skills is licensed Apache-2.0 at the repository root; upstream does not embed an SPDX license identifier in per-skill SKILL.md frontmatter.)

✅ pulumi-component

• Status: Passed
• Findings: 3
• Allowed (not blocking): 1
  • MANIFEST_MISSING_LICENSE (Allowed: pulumi/agent-skills is licensed Apache-2.0 at the repository root; upstream does not embed an SPDX license identifier in per-skill SKILL.md frontmatter.)

✅ pulumi-esc

• Status: Passed
• Findings: 4
• Allowed (not blocking): 1
  • MANIFEST_MISSING_LICENSE (Allowed: pulumi/agent-skills is licensed Apache-2.0 at the repository root; upstream does not embed an SPDX license identifier in per-skill SKILL.md frontmatter.)

✅ pulumi-upgrade-provider

• Status: Passed
• Findings: 8
• Allowed (not blocking): 2
  • MANIFEST_MISSING_LICENSE (Allowed: pulumi/agent-skills is licensed Apache-2.0 at the repository root; upstream does not embed an SPDX license identifier in per-skill SKILL.md frontmatter.)
  • ATR_2026_01610 (Allowed: FP: cisco-ai-skill-scanner matched $(cat /tmp/pr_body.txt) in SKILL.md's documented post-run steps (rule renumbered from ATR_2026_00111 in this scanner version); read-only re-read of a locally-written temp file used to PATCH the PR body via gh api, not privilege escalation or untrusted input.)

✅ upstream-patches

• Status: Passed
• Findings: 3

---

Summary: Scanned 8 skill(s), all passed security checks. ✅

[renovate]

Edited/Blocked Notification

Renovate will not automatically rebase this PR, because it does not recognize the last commit author and assumes somebody else may have edited the PR.

You can manually request rebase by checking the rebase/retry box above.

⚠️ Warning: custom changes will be lost.

[danbarr]

Pushed follow-up commits allowlisting the `skill-security-scan` findings across the three affected skills in this PR:

• `skills/pulumi-upgrade-provider/spec.yaml`: `ATR_2026_01610` matched `$(cat /tmp/pr_body.txt)` in a documented post-run step — the same read-only, self-authored temp-file re-read already justified in this file under the old rule ID `ATR_2026_00111`, just renumbered.
• `skills/provider-upgrade/spec.yaml`: `ATR_2026_00064` matched `chmod +x` in the documented schema-tools install script — standard step making a just-downloaded release binary executable.
• `skills/upstream-patches/spec.yaml`: `ATR_2026_00066` matched `${target_sha}` shell variable interpolation in the documented git rebase workflow — the value comes from the skill's own earlier `git log` lookup, not attacker input.

All three are keyword/pattern false positives, no real secrets or dangerous instructions found. Added allowlist entries with justifications matching each file's existing format.

[danbarr]

Wording correction from independent review on the `ATR_2026_00064` entry I added: "chmod +x" being a standard install step is a valid rebuttal of the privilege-escalation pattern match specifically, but it doesn't independently establish trust in the downloaded schema-tools release binary itself — that's a separate supply-chain question (no checksum/signature verification is documented in this install step). Reworded the entry to scope its claim to the rule it's addressing rather than implying the whole download is vetted.