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.
git clone --depth 1 https://github.com/Fmarzochi/EGCWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/fmarzochi/egc/prp-plan)<a href="https://agentmods.dev/commands/fmarzochi/egc/prp-plan"><img src="https://agentmods.dev/badge/commands/fmarzochi/egc/prp-plan/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/fmarzochi/egc/prp-plan"><img src="https://agentmods.dev/badge/commands/fmarzochi/egc/prp-plan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00012 | $0.00834 |
| Opus 5 | $0.00006 | $0.00417 |
| Sonnet 5 | $0.00002 | $0.00167 |
| Haiku 4.5 | $0.00001 | $0.00083 |
Grade A, and why
prp-plan 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 6d 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementation Plan Builder
Input: $ARGUMENTS
Produce a plan that survives a cold start: a fresh session with zero conversation history must be able to open the plan file and implement the feature without re-researching the codebase. Everything the implementer needs lives inside the plan.
Input resolution
$ARGUMENTS looks like |
Treat as | First action |
|---|---|---|
Path ending in .md |
PRD or phase document | Read it; pick the first unchecked entry under "Implementation Phases" |
| Free text | Feature description | Use the text as the goal statement |
| Blank | Missing input | Ask for a feature description or a PRD path, then stop |
When the source is a PRD, the plan covers one phase, not the whole document. Name the plan after the phase.
Step 1: Reconnaissance
Before writing a single plan line, learn the codebase. Collect and record:
- Conventions: naming, directory layout, error handling style, import style. Cite the config files that enforce them (linter, formatter, tsconfig or equivalent).
- Nearest neighbor: the existing feature most similar to the new one. Note its file paths; the implementation will mirror its shape.
- Test harness: how tests are written and executed here. Copy the exact command that runs them.
- Validation commands: the real lint, typecheck, test, and build commands this repo uses, verified by running them once.
Anything you could not verify goes in the plan under Open risks, never silently guessed.
Step 2: Slice into increments
Cut the work into increments where each one leaves the repo green: compiling, tests passing, feature partially working. Prefer vertical slices (a thin end-to-end path first) over horizontal layers. An increment that cannot be validated on its own is two increments glued together; cut again.
Step 3: Write the plan file
Save to:
PRPs/plans/{kebab-case-feature-name}.plan.md
Create the directory if needed.
Plan format
# Plan: {feature name}
source: {PRD path or "direct request"}
created: {date}
status: pending
## Goal
{Two sentences maximum: what exists when this plan is done, and how we know.}
## Context capsule
Read these before touching anything:
| File | Why it matters |
|---|---|
| {path} | {the pattern or contract it defines} |
Conventions that apply here:
- {convention}: enforced by {config file}
Validation commands (verified working):
- Lint: `{command}`
- Typecheck: `{command}`
- Tests: `{command}`
- Build: `{command}`
## Increments
### [ ] Increment 1: {name}
- Goal: {what works after this increment}
- Files: {paths to create or change}
- Mirror: {existing file whose shape to follow}
- Steps: {short numbered list}
- Validate: `{exact command}` plus {any manual check}
### [ ] Increment 2: {name}
{same structure}
## Open risks
- {unverified assumption or fragile area, and what to do if it bites}
## Done means
- All increments checked
- Full validation suite green
- {feature-specific acceptance check}
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.
- 6d ago First seen · 117 lines · 12 tokens per session scan A 79e97096771b
prp-plan is a command published in the GitHub repository Fmarzochi/EGC (49 stars, last pushed yesterday), licensed Apache-2.0. It adds 12 tokens to every session and 834 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 2026-09-03.
Other commands, from other repositories
demand-discovery
Run automated niche demand discovery research across 7 data sources and 26 categories.
gaai-audit
Show workspace audit log with filters and optional JSON export.
rclone_config_dump
Dump the config file as JSON.
help
Get help with the hookify plugin.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.