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.
git clone --depth 1 https://github.com/Redtropig/harness-anchorWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/redtropig/harness-anchor/test-plan)<a href="https://agentmods.dev/commands/redtropig/harness-anchor/test-plan"><img src="https://agentmods.dev/badge/commands/redtropig/harness-anchor/test-plan.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00063 | $0.00755 |
| Opus 5 | $0.00032 | $0.00378 |
| Sonnet 5 | $0.00013 | $0.00151 |
| Haiku 4.5 | $0.00006 | $0.00076 |
Grade A, and why
test-plan 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 7d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/test-plan
Produce a coverage plan for the active feature: what must be tested, what the suite already
covers, what's missing (including paths the test runner never runs), and the minimal set of tests to
close the gaps. Runs after implementation — it complements /verify (which runs the registered
suite; this finds what's missing from it).
This is the post-implementation, code-aware counterpart to superpowers TDD's pre-implementation
test-first pass. Like /verify, it uses a fresh-context subagent so the analysis isn't biased by
the agent that wrote the code.
Steps
-
Dispatch
coverage-analystvia theTasktool:Task tool with: subagent_type: coverage-analyst prompt: Analyse test coverage for the active feature in feature_list.json (or the feature named in the arguments). Derive obligations from code + spec, diff against the test set and the verified run scope, and recommend a minimal oracle-independent-first test set. Follow the fixed report format in your instructions. Read-only — do not write code or tests.If the user passed a feature id or path as an argument, name it in the prompt.
-
Surface the report verbatim —
### Obligations derived,### Run-scope gaps,### Recommended tests,### Coverage verdict,### Uncertainties. The point of fresh-context analysis is the user sees the analyst's actual words. -
If
### Uncertaintiesis non-empty: ask the user those questions before acting — the analyst escalated because an oracle couldn't be derived. Do not fabricate expected values. -
If verdict is GAPS: offer to act on the recommendations —
- write the recommended tests (oracle-independent first), and/or
- register any orphan binary with the test runner (e.g. add
add_test/gtest_discover_tests),
then run
/verify(and/sanitizefor C/C++) to confirm the now-in-scope tests pass. -
If verdict is COVERED: report it with the
.harness-anchor/coverage-<ts>.mdevidence path — that is the artifact for the "Coverage obligations" criterion inanti-hallucination-gates.
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.
- 7d ago First seen · 64 lines · 63 tokens per session scan A 17af8e0c2a5c
test-plan is a command published in the GitHub repository Redtropig/harness-anchor (13 stars, last pushed 1mo ago), licensed MIT. It adds 63 tokens to every session and 755 once invoked, about $0.0003 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 2026-08-30.
Other commands, from other repositories
test-tdd
Run when user calls /test-tdd. Scans modified files, locates their corresponding unit/integration test suites, and runs them.
complete
Complete a partially implemented feature by filling gaps and ensuring production readiness.
fire-verify-uat
Conversational User Acceptance Testing with automatic parallel diagnosis on failures.
sdd-implement
Execute a task list in dependency order, tests before implementation for each task, halting on any non-parallel failure instead of pressing ahead.
tdd
Test-driven development with TodoWrite planning and comprehensive testing.
modernize-transform
Transform one legacy module to the target stack — idiomatic rewrite with behavior-equivalence tests.