verify

A command that sends a verification agent to review a project task and resolve its response-awareness tags. These tags mark assumptions or decisions that need to be checked during implementation.

In plain words
What is it for?
Use it with a project task to inspect its details, track verification progress, record resolved tags, and update the task status.
Why use it?
It provides a structured review step for checking assumptions and code quality before work is considered ready.

Command for Claude Code

Install

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.

agentmods
npx agentmods add commands/dtannen/pm/verify
Clone the repo
git clone --depth 1 https://github.com/dtannen/pm

Made for: Claude Code.

Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,380 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00011 $0.02380
Opus 5 $0.00005 $0.01190
Sonnet 5 $0.00002 $0.00476
Haiku 4.5 $0.00001 $0.00238

Measured yesterday against content hash e2e04b60aca4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

verify 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 yesterday.

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-plugin/commands/verify.md · 306 lines

How it starts

The opening of the file, as written. The whole thing — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.

RA Tag Verification

Deploy specialized verification agent to systematically resolve Response_Awareness tags, validate implementation assumptions, and ensure production-ready code quality.

Usage

/pm:verify <task-file>

Example: /pm:verify 001-user-auth

Instructions

You are deploying an RA verification agent to resolve all Response_Awareness tags from implementation and ensure code quality through systematic assumption validation.

MCP Integration (preferred)

Drive verification via MCP for locking, status updates, logs, and RA metadata.

  • Full name: mcp__project-manager-mcp__list_tasks
  • Full name: mcp__project-manager-mcp__get_task_details
  • Full name: mcp__project-manager-mcp__acquire_task_lock
  • Full name: mcp__project-manager-mcp__update_task_status
  • Full name: mcp__project-manager-mcp__update_task
  • Full name: mcp__project-manager-mcp__release_task_lock

MCP flow:

  1. Resolve the task by name → ...__list_tasks then ...__get_task_details.
  2. Acquire verification lock → ...__acquire_task_lock(task_id, agent_id) and set status to REVIEW or verifying via ...__update_task_status.
  3. As tags are inventoried and resolved, append structured entries to ra_metadata.verification and log_entry via ...__update_task.
  4. On completion, set DONE (or project’s final state) and release the lock via ...__release_task_lock.

1. Validate Task Implementation

Prefer MCP:

  • Confirm task exists and is ready for verification via ...__get_task_details (status and RA tags).
  • If no RA tags and complexity ≤ 3, you may proceed with Simple verification; otherwise prompt to ensure RA implementation happened.

2. Scan for RA Tags

Read implementation files and identify all RA tags:

  • #COMPLETION_DRIVE_IMPL - Implementation assumptions
  • #COMPLETION_DRIVE_INTEGRATION - Integration assumptions
  • #CONTEXT_DEGRADED - Memory uncertainty
  • #CONTEXT_RECONSTRUCT - Reconstructed details
  • #CARGO_CULT - Pattern-driven code
  • #PATTERN_MOMENTUM - Over-implementation
  • #ASSOCIATIVE_GENERATION - "Feels right" additions
  • #PATTERN_CONFLICT - Conflicting approaches
  • #TRAINING_CONTRADICTION - Training conflicts
  • #PARADIGM_CLASH - Architecture conflicts
  • #BEST_PRACTICE_TENSION - Competing practices
  • #SUGGEST_ERROR_HANDLING - Error handling suggestions
  • #SUGGEST_EDGE_CASE - Edge case suggestions
  • #SUGGEST_VALIDATION - Validation suggestions
  • #SUGGEST_CLEANUP - Cleanup suggestions
  • #SUGGEST_DEFENSIVE - Defensive programming suggestions

Read the full file on GitHub · 306 lines

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.

  1. yesterday First seen · 306 lines · 11 tokens per session scan A e2e04b60aca4

Subscribe to this mod's changes

verify is a command published in the GitHub repository dtannen/pm (0 stars, last pushed 10mo ago), licensed MIT. It adds 11 tokens to every session and 2,380 once invoked, about $0.0001 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.