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 skills/greppable/greppable-cc-plugin/managing-gdl-rulesnpx skills add greppable/greppable-cc-plugin --skill managing-gdl-rulesgit clone --depth 1 https://github.com/greppable/greppable-cc-pluginWrote 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/skills/greppable/greppable-cc-plugin/managing-gdl-rules)<a href="https://agentmods.dev/skills/greppable/greppable-cc-plugin/managing-gdl-rules"><img src="https://agentmods.dev/badge/skills/greppable/greppable-cc-plugin/managing-gdl-rules.svg" alt="Measured on agentmods" 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 | $0.00100 | $0.00758 |
| Opus 5 | $0.00050 | $0.00379 |
| Sonnet 5 | $0.00020 | $0.00152 |
| Haiku 4.5 | $0.00010 | $0.00076 |
Grade A, and why
managing-gdl-rules 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 4d 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GDL Rules Reference
Current Rules
!bash -c 'if [ -f rules.gdl ]; then echo "From rules.gdl:"; echo ""; grep "^@rule" rules.gdl 2>/dev/null; else echo "No rules.gdl found in project root."; fi'
What Makes a Rule Worth Writing
A rule captures a project-specific convention with codebase evidence. Not a generic best practice.
| Good rule (project-specific) | Bad rule (generic) |
|---|---|
| "API routes must validate input with zod schemas" | "Use descriptive variable names" |
| "Never modify existing migrations — create new ones" | "Write clean code" |
| "React components use PascalCase filenames" | "Follow naming conventions" |
The test: Can you point to 3+ files that follow this convention? If yes, it's a rule. If it's advice you'd give on any project, it's not.
When to Write vs Observe
- Write a @rule when: you've seen the convention in 3+ files, or the user explicitly confirms it
- Write a @memory observation when: you've noticed a pattern but aren't sure it's intentional
- Don't write anything when: it's a language-level best practice that any linter would catch
Format
@rule|scope:GLOB|severity:LEVEL|desc:TEXT
| Field | Values | Meaning |
|---|---|---|
scope |
Glob pattern (*.ts, src/api/**, **/*.test.*) |
Which files this rule applies to |
severity |
error |
Will break things if violated — wrong behavior, data loss, security |
warn |
Convention that should be followed — consistency, maintainability | |
info |
Team preference — style, approach | |
desc |
Free text | What the rule requires (imperative form) |
Tool Function
source "${CLAUDE_PLUGIN_ROOT}/scripts/gdl-tools.sh"
# Check which rules apply to a file
gdl_rules_for_file "src/api/routes.ts" rules.gdl
Grep Patterns
# All rules
grep "^@rule" rules.gdl
# Rules for a specific scope
grep "^@rule|scope:src/api" rules.gdl
# All error-severity rules
grep "severity:error" rules.gdl
# Rules matching a file (fuzzy)
grep "^@rule" rules.gdl | grep "\.ts"
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.
- 4d ago First seen · 81 lines · 100 tokens per session scan A 39f4c6a575df
managing-gdl-rules is a skill published in the GitHub repository greppable/greppable-cc-plugin (3 stars, last pushed 4mo ago), licensed MIT. It adds 100 tokens to every session and 758 once invoked, about $0.0005 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 skills, from other repositories
ygrep
IMPORTANT: Try ygrep FIRST for all code and file searches before using Grep, Glob, or Task explore. ygrep uses a pre-built full-text index and returns results in milliseconds. Fall back to built-in tools only if ygrep returns no results.
semantic-search
Code search for this repo — pick the right Beacon tool for the question: exact symbol lookup, reference tracing, file outline, or semantic search.
ripgrep
Use when the user needs text or regex content search with ripgrep: composing rg commands, choosing flags, glob/type filtering, multiline or PCRE2 searches, pipeline output, grep-to-rg migration, or diagnosing why rg missed a file (gitignore, hidden, binary defaults). Not for syntax-aware structural queries (ast-grep)…
writing-docs
Write or restructure documentation that agents and people actually read — choosing which of the four directories it belongs in (how-to, reference, decisions, exec-plans), knowing what belongs in a failure message instead of a file, and keeping the routing table honest. Use this whenever writing a doc, a runbook, a…
repo-index
Give a repository a code-and-docs graph an agent can query — tree-sitter symbols, import and call edges, doc-to-code edges, ranked by personalized PageRank from whatever the current task touches. Use this when an agent cannot find the relevant code in a large repo, when grep returns hundreds of hits or none, when…
writing-checks
Build the machinery that enforces a convention instead of documenting it — PreToolUse guards that block an action before it runs, CI gates that judge the worktree, selftests that prove a check can turn red, and structural tests that enforce layering. Use this whenever a rule keeps getting violated despite being…