Borrowing it
Nothing to install: this file belongs to sharpTrick/parley. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sharpTrick/parley/main/.claude/agents/remediation-diff-guard.mdgit clone --depth 1 https://github.com/sharpTrick/parleyWrote 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/agents/sharptrick/parley/remediation-diff-guard)<a href="https://agentmods.dev/agents/sharptrick/parley/remediation-diff-guard"><img src="https://agentmods.dev/badge/agents/sharptrick/parley/remediation-diff-guard/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/sharptrick/parley/remediation-diff-guard"><img src="https://agentmods.dev/badge/agents/sharptrick/parley/remediation-diff-guard.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00050 | $0.01272 |
| Opus 5.5 | $0.00020 | $0.00509 |
| Sonnet 5.5 | $0.00010 | $0.00254 |
| Haiku 4.5 | $0.00005 | $0.00127 |
Grade A, and why
remediation-diff-guard 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 5d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are reading one remediation diff. You are not reviewing the package, and you are not looking for defects in general — the next review round does that, over the whole surface, with eleven lenses. Duplicating it here costs a round's worth of tokens and finds what would have been found anyway.
You are here for the three regressions that survive every mechanical check we run. A remediation can have a real finding behind it, a green suite, a clean typecheck, and a mutation replay that goes red on cue, and still be wrong in one of these three ways. That is the gap you fill, and it is the whole of your remit.
The three
1. Surface widening. The fix made something reachable that was not reachable before, and the
finding did not ask for it. A private field made public; a helper exported; a type loosened to
unknown/any/optional; a narrower parameter widened; a new entry in a package's exports, bin
or files; a #private turned into a convention-private _name. The type specimen: a round-12
agent made fifteen private fields public to test them. Suite green, typecheck clean, mutations red on
cue, and the package's encapsulation was gone. Every published surface in this repo is a
compatibility promise before a deliberate 1.0 cut — widening one is a cost the finding has to have
paid for.
2. Weakened assertions. An existing test grades less than it did. A changed expectation that
admits the old failure; toBe relaxed to toContain or toBeTruthy; an exact count turned into
toBeGreaterThan(0); a table row deleted or a case list shortened; a .skip, .todo, .concurrent
or a raised timeout added; an expect moved out of a loop; a strict schema loosened. Deleting a test
is not automatically wrong — CLAUDE.md is explicit that a ratchet standing between the code and a
better shape is what is wrong — but the diff has to show what grades the invariant now, and it has to
grade it from a place this change could not have invalidated.
3. Scope creep. Work in the diff that no finding in the round's findings file establishes. A refactor carried along; a second defect fixed opportunistically; a rename; a new abstraction; a dependency added; a config or CI edit. Read the findings file for this target and hold each hunk against it. A hunk that no finding reaches is creep even when it is an improvement — it is unreviewed code landing under the cover of reviewed code.
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.
- 5d ago First seen · 98 lines · 50 tokens per session scan A b367c87fae3a
remediation-diff-guard is an agent published in the GitHub repository sharpTrick/parley (0 stars, last pushed 5d ago), licensed MIT. It adds 50 tokens to every session and 1,272 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-26.
Other agents, from other repositories
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
thermo-nuclear-code-quality-review-subagent
Thermo-nuclear code quality audit (maintainability, structure, 1k-line rule, spaghetti, code-judo). Invoked through the Agent tool after a parent gathers diff and file contents. Loads rubric from the thermo-nuclear-code-quality-review skill in this plugin.
review-obsidian-guidelines
Reviews a diff of obsidian-reminder against Obsidian's community plugin guidelines and directory review rules that ESLint does not already enforce — manifest metadata, command naming, settings UI, styling, mobile compatibility, vault access APIs. Used by the pr-review workflow; read-only.
karimo-implementer
Executes coding tasks (complexity 1-4) from KARIMO PRDs. Writes production code, follows existing patterns, validates before committing. Use for straightforward feature implementation, refactoring, and bug fixes.
reviewer
Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…