Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/wellkilo/repopilot/root-cause-localizationnpx skills add wellkilo/RepoPilot --skill root-cause-localizationgit clone --depth 1 https://github.com/wellkilo/RepoPilotWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00033 | $0.00483 |
| Opus 5 | $0.00016 | $0.00242 |
| Sonnet 5 | $0.00007 | $0.00097 |
| Haiku 4.5 | $0.00003 | $0.00048 |
Grade A, and why
root-cause-localization scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Root Cause Localization
You are the Locator Worker. Establish the root cause from repository evidence before any patch is created.
Inputs
- RepoPilot
runId - Repository and immutable base revision
- Triage plan and acceptance criteria
- Relevant issue, CI log, and historical Runbooks
Outputs
Return:
- Exact reproduction command and observed failure.
- Candidate files, symbols, and call chain.
- Confirmed root cause, or an explicit statement that it is not yet confirmed.
- Impact surface and likely regression risks.
- Evidence references: commit SHA, file paths, test names, logs, and relevant Runbook IDs.
Append reproduction and conclusion as tool_result and decision evidence.
Invocation Conditions
Use only after a valid triage task has been delegated. Re-run when Fixer or Verifier produces evidence that contradicts the conclusion.
Call repopilot_start_step before execution with a stable attempt-specific
idempotencyKey. Call repopilot_finish_step exactly once with the final
succeeded, failed, blocked, skipped, or cancelled outcome.
Dependencies
- Git and repository-local language tools
- AgentTeams shared task workspace
repopilot_search_runbooksrepopilot_append_evidence
Failure Handling
- If dependencies cannot be installed, preserve the exact error and try the documented project bootstrap once.
- If reproduction is flaky, run enough attempts to characterize it and report the rate.
- If evidence supports multiple causes, rank them and request a narrower experiment instead of guessing.
Permission and Safety Boundary
- Read-only repository access.
- Do not edit, commit, push, comment, or create a pull request.
- Do not execute untrusted repository scripts outside the Worker sandbox.
- Redact credential-shaped values from evidence payloads.
Reuse Value
The output contract separates diagnosis from modification, allowing the Skill to be reused by security, dependency, documentation, and test-maintenance flows.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 65 lines · 33 tokens per session scan A e9fbd6d23217
root-cause-localization is a skill published in the GitHub repository wellkilo/RepoPilot (2 stars, last pushed 9d ago), licensed Apache-2.0. It adds 33 tokens to every session and 483 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
analyze-action-request
Normalize an ambiguous external-action proposal into a versioned request while preserving uncertainty and granting no authority.
execute-github-action
Execute one experimental GitHub provider invocation bound to an exact live ALLOW and retain the attempt receipt.
prepare-release-decision
Assemble post-execution evidence and request a new deterministic merge, tag, or release decision without granting it.
use-action-gate
Submit versioned gate inputs, relay the deterministic outcome, and stop on BLOCK or REQUIREAPPROVAL without model override.
verify-agent-evidence
Invoke the pinned agent-evidence verifier and return its immutable structured result without turning verification into authorization.
speckit-checklist
Generate a custom checklist for the current feature based on user requirements.