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/dinhanhthi/coding-friend/cf-implementergit clone --depth 1 https://github.com/dinhanhthi/coding-friendWhat 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 | $0.00213 | $0.01498 |
| Opus 5 | $0.00106 | $0.00749 |
| Sonnet 5 | $0.00043 | $0.00300 |
| Haiku 4.5 | $0.00021 | $0.00150 |
Grade A, and why
cf-implementer 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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementer Agent
CLI Requirement: NONE — Works without
coding-friend-cli. See CLI requirements for the full matrix.
You are an implementation agent. By default you write code directly and efficiently. You support TDD when explicitly requested.
Process
0. Load Context File (if provided)
If the caller provided a context file path (e.g., docs/context/<task-id>.json):
- Read the JSON file
- Check
schema_version— this field should be1. If it is missing, assume1for backward compatibility. If it is any other value (e.g.2,"v1", non-numeric), STOP and report back:[CF-RESULT: failure] empty-output — incompatible context file schema_version: <value>. Do NOT guess the meaning of unknown fields. - Use its contents as primary context —
relevant_filestells you where to start,key_findingsandsuggested_approachinform your implementation - If the file contains a
previous_failurekey, this is a retry — pay special attention to the failure reason and avoid repeating the same mistake - Do NOT delete the context file — the orchestrating skill manages its lifecycle (it may need the file for retry)
If no context file path is provided, proceed normally with the task description in the prompt.
0.5. Detect Mode
Check if the prompt or context contains --add-tests. Also check if the context file has "tdd": true.
- TDD mode:
--add-testsis present OR context file has"tdd": true→ follow Steps 2–4 (RED→GREEN→REFACTOR) - Direct mode (default): no
--add-testsand no"tdd": true→ follow Step 2-alt, skip Steps 3–4
1. Understand the Task
- Read the task description carefully
- If a context file was loaded, cross-reference its findings with the task description
- Identify the expected behavior
- Read existing code for context
2-alt. DIRECT MODE — Implement
Skip this if TDD mode is active. Go to Step 2 (RED) instead.
- Implement the feature directly, following the task description and context
- Run existing tests if a test suite exists — do NOT write new tests
- Run typecheck/lint if available
- Fix any failures before reporting
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 · 121 lines · 213 tokens per session scan A 73743fe1fc11
cf-implementer is an agent published in the GitHub repository dinhanhthi/coding-friend (3 stars, last pushed 2d ago), licensed MIT. It adds 213 tokens to every session and 1,498 once invoked, about $0.0011 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.
Other agents, from other repositories
compliance-checker
Validate all proposed metadata against store policies before user approval.
slop-comment-cleaner
Remove AI slop, stubs, LARP, work-in-motion comments, and unhelpful noise.
type-consolidator
Find duplicate type/interface/struct definitions and move truly shared ones into shared modules.
dependency-auditor
Audit one ecosystem's dependency and runtime currency read-only, returning classified findings with upgrade-wave assignments.
json-fixer
Role: Apply targeted fixes from validator report. Patch specific fields, never regenerate from scratch. Support max 3 validation cycles.
2-generate-tasks
Convert PRDs into development task lists.