A guide for adding a deterministic triage check to whodecided, a tool that reviews decisions and code changes. The check reads structured project data and reports a result across the tool's review, board, and report views.

Use it to add risk factors, links between decisions and code changes, or coverage checks. It covers the check's code, tests, data wiring, and display.

It prevents the same check from being implemented differently in several places. It also keeps checks independent of language-specific wording and large language models.

How zhu-kai/whodecided configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to zhu-kai/whodecided. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/zhu-kai/whodecided/main/.claude/skills/add-signal/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/zhu-kai/whodecided

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/zhu-kai/whodecided/add-signal/github.svg)](https://agentmods.dev/skills/zhu-kai/whodecided/add-signal)

Your own site

<a href="https://agentmods.dev/skills/zhu-kai/whodecided/add-signal"><img src="https://agentmods.dev/badge/skills/zhu-kai/whodecided/add-signal/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/zhu-kai/whodecided/add-signal"><img src="https://agentmods.dev/badge/skills/zhu-kai/whodecided/add-signal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 51 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 416 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00051	$0.00416
Opus 5.5	$0.00020	$0.00166
Sonnet 5.5	$0.00010	$0.00083
Haiku 4.5	$0.00005	$0.00042

Measured 1mo ago against content hash 0825c0cdc380, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

add-signal 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 1mo 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.

.claude/skills/add-signal/SKILL.md · 17 lines

What it actually says

A signal is a deterministic judgement over data already on hand (ledger entries, branch diff, traces). Adding one:

Pure function in src/checks/ - data in, struct out, unit-tested directly. Never call git, fs, or an LLM from checks; the impure gathering lives in src/git.ts and src/commands/signals.ts.
Language rule. Ledger prose is in the user's language, so a signal must not depend on matching prose keywords. Semantics come from structured fields - if the meaning is only in the prose, add a distiller tag (tags in src/ledger/schema.ts + the tags contract in src/distill/prompt.ts) and match on the tag. English keyword regexes are acceptable only as a fallback for pre-tag entries.
Wire once in src/commands/signals.ts (analyzeBranch for per-branch data, decisionSignals for per-decision) - review, board, and report all consume from there. Do not compute signals inside a command.
Surface it: src/render/html.ts (WorkbenchEntry field + drawer/attention/triage as fits) and src/render/md.ts if it belongs in the PR report. Negative/actionable signals go in the drawer's needs-attention block; healthy states stay out of the way.

Scoring conventions in checks/risk.ts: silent 35 / hard-to-reverse 25 / assumption 20; diff churn, relative churn ("rewrote N% of a file", new files excluded), and scatter (entropy) top it up. High >= 60, medium >= 32. Keep new factors proportionate.

Changes

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.

1mo ago First seen · 17 lines · 51 tokens per session scan A 0825c0cdc380

Subscribe to this mod's changes

add-signal is a skill published in the GitHub repository zhu-kai/whodecided (22 stars, last pushed 3mo ago), licensed MIT. It adds 51 tokens to every session and 416 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 2026-08-30.

Claude Code Code review LLM ai-agents cli developer-tools

hypotheses

Turns a research direction into testable hypotheses with predictions, competing explanations and the experiments that discriminate between them, including the design (controls, randomization, blocking), sample size or seed count, the pre-specified analysis, and the result that would falsify each hypothesis. Use when a…

synthetic-sciences/openscience · 99 tokens

crit-cli

Use when an agent needs to author or reply to crit inline comments programmatically (including multi-agent workflows commenting on shared code/plans/docs/proposals), publish or unpublish a crit review with crit share, sync a crit review to or from a GitHub PR or GitLab MR, or read/interpret a crit review JSON file.…

tomasz-tomczyk/crit · 110 tokens

crit

Review code changes, a plan, a live page (running dev server), or a local HTML file with Crit inline comments and structured human feedback. Use only when the user explicitly invokes /crit or directly asks to use Crit; a generic review request does not count.

tomasz-tomczyk/crit · 55 tokens

crit-story

Author a crit story and continue the interactive review loop only when the user explicitly invokes crit-story or directly asks you to generate a crit story. Do not infer this skill from generic review, PR, or diff-review requests.

tomasz-tomczyk/crit · 47 tokens

crossreview

Run when the user invokes /crossreview (or /crossreview:setup to choose the reviewers) or asks for a cross-review, a quorum review or a second opinion from other agents or models: fan a code review (uncommitted work, a branch, a commit range, files, a plan or a document) out to several console code agents (Claude…

coddy-project/coddy-agent · 204 tokens

review-harness

Use when reviewing implementation work that ran through a /tmp Pi task harness. Checks contract drift, unsafe shortcuts, validation evidence, and status honesty.

spences10/my-pi · 33 tokens

whodecided: Skill for Claude Code