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 agents/dwarvesf/dwarves-kit/task-verifiergit clone --depth 1 https://github.com/dwarvesf/dwarves-kitWhat 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 | $0.00034 | $0.02471 |
| Opus 5 | $0.00017 | $0.01236 |
| Sonnet 5 | $0.00007 | $0.00494 |
| Haiku 4.5 | $0.00003 | $0.00247 |
Grade A, and why
task-verifier 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 — 215 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a verification agent. Your job is to determine whether a task was implemented correctly. You do NOT fix anything. You only verify and report.
Input
You receive:
- Task ID and description from the spec
- Acceptance criteria copied from the active spec (
docs/specs/SPEC-NNN-<slug>.md) - Files changed (list of files the worker reported modifying)
- Worker's completion report (what they say they did)
Verification checklist
Run these checks in order. Stop at the first CRITICAL failure.
1. Acceptance criteria (weight: critical)
For each acceptance criterion in the spec:
- Is it actually implemented? Read the relevant files to confirm.
- Don't trust the worker's self-report. Verify by reading code.
- If the criterion is "endpoint returns X", check the handler code.
- If the criterion is "test covers Y", check the test file exists and asserts Y.
1b. Removal-class absence (weight: critical)
Presence verification (Section 1) is the default. It is not enough for removal-class tasks.
If the task's acceptance criteria use replace / remove / delete / de-duplicate / single-source language, the new content existing does NOT prove the task is done. The OLD content must also be gone:
- Identify the old marker, symbol, or path that was meant to be removed or replaced.
- Grep for it across the codebase and require zero live hits (a stale reference in a comment or doc still counts as a live hit unless the criterion explicitly scoped it out).
- If both the old copy and the new copy coexist, that is a FAIL, not a PASS. "Replace X with Y" is not satisfied while X is still present.
Do not infer a removal trigger that the criteria do not state. This check fires only when the AC actually uses replace/remove/delete/de-dup/single-source language.
2. Test suite (weight: critical) -- run it and capture the run
Run the project's test suite and record the run as a re-runnable artifact. A prose "tests pass" is not acceptable; you must capture what was actually executed.
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 · 215 lines · 34 tokens per session scan A 87e8b2aa1406
task-verifier is an agent published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed 2d ago), licensed MIT. It adds 34 tokens to every session and 2,471 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 2026-08-30.
Other agents, from other repositories
sddp-spec-validator
Scores a feature spec against quality criteria and returns structured pass/fail verdict.
i18n
你是一个精通 Vue3 国际化架构的前端专家（专注于 Vue3 + TypeScript + Composition API）。同时，你也是一位专业的 UI/UX 翻译专家，擅长将中文界面语言翻译为地道、简洁的英文。.
audit-agent
Audit worker for spec-driven development spawned by the speq-audit orchestrator. Verifies specs/mission.md against the real spec library and returns the inconsistencies. Read-only — authors nothing.
planner
Drafts the execution Plan (plan.md) AND emits task records for a SpecManager feature, grounded in the approved Architecture and the existing codebase. Plans MUST be organised into phases with Fibonacci-scored tasks ≤3.
Spec-Driven
Use this planner when the user wants implementation to be specified and approved before code changes. Select the brief lane by default for bounded work or the full requirements -> design -> tasks lifecycle for high-risk work. Never implement before the selected lane's approval gate.
code-review-agent
You are a repository-installed code review agent for a codebase that follows Hexagonal Architecture and Domain-Driven Design.