fix(skills): make helper references resolvable at first mention

[thewrz]

This was written agentically; verify its assertions:

Why

Bare helper names and ambiguous library references made agents guess paths or interfaces, sometimes turning failed discovery into an apparent empty issue selection.

What

Require resolvable $agentkit-relative first mentions for shipped helpers and add regression coverage. Clarify sourced libraries and the real contract-cache CLI exception, show the provider-config --repo-root interface, and require degraded selection to report unknown eligibility.

Decisions

Inventory shipped helpers so external and example scripts are exempt. Preserve skill-size limits by condensing adjacent prose and updating equivalent exact-wording assertions. The documentation changes and PR #604 merge cleanly: git merge-tree --write-tree returned integration tree 2eb2e994d78520ee6aeff7d6ba2177e3bb7eb336 with exit 0; their changed passages are disjoint.
base=origin/main
files=10
total.insertions=200
total.deletions=115
total.lines=315
operational.files=10
operational.insertions=200
operational.deletions=115
operational.lines=315
generated.files=0
generated.insertions=0
generated.deletions=0
generated.lines=0
lockfile.files=0
lockfile.insertions=0
lockfile.deletions=0
lockfile.lines=0
fixture.files=0
fixture.insertions=0
fixture.deletions=0
fixture.lines=0
non_operational.files=0
non_operational.insertions=0
non_operational.deletions=0
non_operational.lines=0

Testing

• Regression red: four helper-reference and two documentation-contract failures before implementation.
• Focused helper-refs and skills-contract checks: 27 and 174 assertions passed.
• agent-run.sh --cmd test: all 107 suites and shared shell, recipe, path, and skill-size gates passed (ALL GREEN).

🤖 Co-authored by Codex GPT-6 Astra.

Closes #602

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Team

Run ID: cf9474d1-1eae-4cad-b34d-bd7d95d2d17a

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

_{Comment @coderabbitai help to get the list of available commands.}

[thewrz]

This was written agentically; verify its assertions:

Adversarial review receipt

• Reviewer: provider=anthropic; model=claude-fable-5-1; effort=high; mode=cross-provider (reason: Explicit invocation override; Fable 5.1 high)
• Counts: P1=0; P2=0; total=0
• Reviewed head: 129c7fd
• Diff payload: wrzonance/agent-kit:605:b41b9d1d710f5a6f71e6f77da5d258399e5fbb85c0a74c2829cde81696ec17f9
• Confirmed finding: none confirmed

🤖 Co-authored by Codex GPT-6.

[thewrz]

This was written agentically; verify its assertions:

Review ledger

Machine-readable record of every review already performed on this PR.

{
  "version": 1,
  "pr": 605,
  "repo": "wrzonance/agent-kit",
  "reviews": [
    {
      "kind": "adversarial",
      "provider": "anthropic",
      "model": "claude-fable-5-1",
      "effort": "high",
      "mode": "cross-provider",
      "harness": "codex",
      "head_sha": "129c7fdc78afd2e606d63937e3dc29903805d416",
      "covered_heads": [
        "129c7fdc78afd2e606d63937e3dc29903805d416"
      ],
      "diff_payload": "wrzonance/agent-kit:605:b41b9d1d710f5a6f71e6f77da5d258399e5fbb85c0a74c2829cde81696ec17f9",
      "counts": {
        "p1": 0,
        "p2": 0
      },
      "reviewed_at": "2026-09-05T05:48:15Z"
    }
  ]
}

🤖 Co-authored by Codex GPT-6.