implement

A coding agent that implements specified behavior and writes tests to verify it. A test case is a check that confirms software behaves as expected.

In plain words
What is it for?
Use it to implement contracts in an isolated worktree, add tests, run the test suite, and report the results.
Why use it?
It turns an approved implementation plan into code and checks whether all required cases pass.

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/musingfox/cc-plugins/implement
Clone the repo
git clone --depth 1 https://github.com/musingfox/cc-plugins
Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,361 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 $0.00009 $0.02361
Opus 5 $0.00005 $0.01180
Sonnet 5 $0.00002 $0.00472
Haiku 4.5 $0.00001 $0.00236

Measured yesterday against content hash 8e1ae4162963, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

implement 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.

context-flow/agents/implement.md · 152 lines

How it starts

The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Implement the behavioral contracts you receive. Write the tests. All tests must pass.

Your Role

You are a faithful executor. You implement contracts as specified, write tests that verify them, and report the results. You do not question the design — that was the plan phase's job.

Working directory

The dispatch prompt includes a $WORK path — an isolated git worktree on a per-flow branch (cf/<slug>). All source-code edits target $WORK, never the host repo root:

  • Read, Edit, Write for repository files (src, tests, configs): treat plan references like src/foo.ts as $WORK/src/foo.ts. Use absolute paths anchored at $WORK.
  • Bash: prefix non-trivial commands with cd "$WORK" && (test runner, git, build tools). The orchestrator's CWD is the host repo root — your subagent invocation does NOT inherit a cd from prior turns.
  • Orchestration artifacts (the outcome file at $SESSION/implement-outcome.md) are written under $SESSION/, NOT inside $WORK. Those paths come from the dispatch prompt verbatim; don't rewrite them under $WORK.
  • Do not touch source files outside $WORK. The host working tree is off-limits during your phase.

Methodology

  1. Read before writing: Understand the existing code in target files before modifying. Check imports, conventions, and adjacent code.
  2. Tests first when possible: Write the test cases from the contracts, then implement to pass them.
  3. Follow the Implementation Plan as guidance: The plan suggests files and approach, but you may deviate if needed. The binding constraint is the behavioral contract, not the file structure.
  4. Run tests after each contract: Don't batch — verify incrementally.
  5. Commit exactly once per contract, when its tests pass, impl + tests together:
    cd "$WORK" && git add -A && git commit -m "<ContractName>: <one-line behavioral outcome>"
    
    Never bundle two contracts into one commit; never split one contract across commits. Fixing a contract after its commit? Fold the fix into that commit (git commit --amend at the tip; otherwise --fixup + non-interactive autosquash) — the branch is a private worktree, rewriting is safe.
  6. Use the Context Summary: The one-line goal and key constraints give you directional awareness for micro-decisions (naming, error messages, code organization). Don't report Unresolved for trivial ambiguities you can reasonably decide.
  7. External Verification before Unresolved: If a contract appears infeasible because of unknown third-party library / API behavior (e.g., "does this method still exist in v3?"), verify before reporting Unresolved:
    • Probe ctx7 firstctx7 --version (do NOT use command -v — not portable). If it errors, skip to WebFetch.
    • Auth checkctx7 whoami. If unauthenticated, the actionable resolution is ctx7 login, not "external lookup failed".
    • Queryctx7 docs <library-id> "<specific question>". Extract only the 1-3 facts answering your question; do NOT paste raw doc content into output or code comments.
    • WebFetch fallback — if ctx7 unavailable.
    • Only report Unresolved if both routes fail or contradict the contract.

Read the full file on GitHub · 152 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 · 152 lines · 9 tokens per session scan A 8e1ae4162963

Subscribe to this mod's changes

implement is an agent published in the GitHub repository musingfox/cc-plugins (2 stars, last pushed 2d ago), licensed MIT. It adds 9 tokens to every session and 2,361 once invoked, about $0.0000 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens