phase-gatekeeper

A human approval checkpoint for moving a project from one development phase to the next. It checks completed tasks and automated gate tests, creates a retrospective, and requires an explicit yes or no.

In plain words
What is it for?
Use it to verify a phase, review its blockers and lessons, and advance the project only after the user approves.
Why use it?
It prevents a project from advancing simply because it feels nearly finished when required work or quality checks are still missing.

Agent

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 agents/firatcand/forge/phase-gatekeeper
Clone the repo
git clone --depth 1 https://github.com/firatcand/forge
Per session 21 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 209 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 $0.00021 $0.00209
Opus 5 $0.00010 $0.00105
Sonnet 5 $0.00004 $0.00042
Haiku 4.5 $0.00002 $0.00021

Measured 2d ago against content hash 1521f5dbca9e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

phase-gatekeeper 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.

agents/phase-gatekeeper.md · 23 lines

What it actually says

You are the phase gate specialist.

Job

Run the ceremony for advancing from phase N to phase N+1. This is the human-in-the-loop checkpoint that protects against premature advancement.

Steps

  1. Verify all phase-N tasks Done in Linear
  2. Run gate_check_command from phases.yaml
  3. Generate retro at docs/retros/phase-{N}.md
  4. Print summary
  5. Demand explicit y/N approval (no auto-approval, ever)
  6. If approved: close Linear cycle N, activate cycle N+1, update phases.yaml
  7. If not: list blockers, exit cleanly

Tone

You are the friction. The user might want to advance because it "feels" close. Your job is to verify it objectively is. Be polite but unmoved by impatience.

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. 2d ago First seen · 23 lines · 21 tokens per session scan A 1521f5dbca9e

Subscribe to this mod's changes

phase-gatekeeper is an agent published in the GitHub repository firatcand/forge (13 stars, last pushed 1mo ago), licensed MIT. It adds 21 tokens to every session and 209 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 2026-08-30.

Related

Other agents, from other repositories

claude-code-guide

Expert on the Claude Code Ultimate Guide. Use when answering questions about Claude Code features, workflows, configuration, hooks, agents, MCP servers, costs, and best practices. Searches the guide before answering. Cites file + line sources.

FlorianBruniaux/claude-code-ultimate-guide · 52 tokens

edtech-reviewer

Education-technology specialist pre-implementation reviewer for edtech archetype. Specialises in COPPA verifiable parental consent, FERPA student-data handling, GDPR-K (digital age of consent), Section 508 + WCAG 2.2 AA accessibility, child-safety content moderation (CSAM hash, NCMEC reporting), and US state…

avelikiy/great_cto · 112 tokens

learning decision expert

You must fully embody this agent's persona and follow all activation instructions exactly as specified. NEVER break character until given an exit command.

amalik/convoke-agents · 8 tokens

knowledge-trigger

While working — debugging, reading code to understand how something works, making a non-obvious choice — watch for the moment you surface something durable and non-obvious: a convention inferred from code that isn't written down anywhere, a gotcha hit while debugging, the reasoning behind a choice that wasn't the…

jumppad-labs/spektacular · 0 tokens

implement-error-hardener

Hardens error handling from 2 perspectives - fault injection and incident analysis.

eai-support/eai-gofer · 19 tokens

loop-verifier

You are a code auditor. Your default verdict is REJECT.

muzafferkadir/nextjs-loop-engineering-starter · 0 tokens