greplm-search

A code-search helper for exploring an unfamiliar codebase and finding where code is defined, used, or connected. It can build an index and provide code-aware search and references.

In plain words
What is it for?
Use it to locate symbols and implementations, follow references and call relationships, inspect focused code sections, and assess the impact of changes.
Why use it?
It reduces the time spent opening many files and manually tracing how a piece of code works.

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/khaledsmq/greplm/cursor
Clone the repo
git clone --depth 1 https://github.com/KhaledSMQ/greplm
Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,066 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. Scan, not verified.
Origin 97% copy Near-identical to another mod 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 $0.00042 $0.01066
Opus 5 $0.00021 $0.00533
Sonnet 5 $0.00008 $0.00213
Haiku 4.5 $0.00004 $0.00107

Measured 2d ago against content hash e4e28281be0e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

greplm-search scanned grade C with 2 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

curl -fsSL https://raw.githubusercontent.com/KhaledSMQ/greplm/main/install.sh | sh

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -fsSL https://raw.githubusercontent.com/KhaledSMQ/greplm/main/install.sh | sh
Origin

This is a copy

97% identical to greplm-search — 7 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

crates/greplm-cli/agents/cursor.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Use the greplm MCP server when it is configured (greplm mcp config → paste into .cursor/mcp.json). Call build_context first, then search_code, find_symbol, impact_of, and the other MCP tools — they return compact, jump-ready locations instead of whole files.

If MCP is not set up yet, use the greplm CLI the same way — faster than raw grep, with real code intelligence (call graph, typed go-to-definition, structural search, git history).

greplm index                          # build or refresh the index

# Find code
greplm search "authentication flow"
greplm search "SegmentWriter" --word  # whole-identifier match
greplm search -e 'fn .*candidates' --lang rust --limit 20
greplm search "needle" --exhaustive       # every match: no ranking/limit/per-file cap (grep parity)
greplm symbols extract --exact        # symbol / definition lookup
greplm outline crates/greplm-core/src/trigram.rs
greplm snippet crates/greplm-core/src/trigram.rs 15 25 --context 3

# Code intelligence
greplm pack "how does incremental indexing work" --budget 8000  # task -> ranked, budgeted context
greplm def <file> <line> <col>   # typed go-to-definition (file line col)
greplm callers references            # who calls this symbol
greplm callees merge_segments        # what this symbol calls
greplm impact add_doc --depth 3      # blast radius: what breaks if I change this
greplm xref SegmentWriter            # resolved references (defs + calls + imports)
greplm ast 'fn $NAME() {}' --lang rust   # structural / AST search (or a tree-sitter query)

# Git time-travel
greplm blame <file> <line>
greplm history references            # commits that touched a symbol
greplm changed main                  # files + symbols changed since a revision

greplm summary

Pass -C ./my-project to search another directory. Add --json for machine-readable output.

For repeated queries in an agent loop, run greplm serve in the background to keep the index hot; queries then route to the daemon automatically.

If greplm is not on $PATH, install with:

curl -fsSL https://raw.githubusercontent.com/KhaledSMQ/greplm/main/install.sh | sh

Workflow

  1. Run greplm index before searching a project (incremental by default). indexed 0 files means nothing changed since the last run — not an empty index; confirm the index is populated with greplm status/greplm summary (check files/doc_count) before falling back to grep.
  2. Start a task with greplm pack "<task>" --budget N to load exactly the relevant code instead of reading whole files.
  3. Before editing a symbol, run greplm impact <symbol> to see the blast radius, plus greplm callers/greplm callees to map the call graph.
  4. Use greplm def <file> <line> <col> for typed go-to-definition and greplm xref <symbol> for resolved references (definitions, calls, imports).
  5. Use greplm ast '<pattern>' --lang <lang> for structural matches regex can't express.
  6. Use greplm blame/greplm history/greplm changed to understand how and why code evolved.
  7. For "find every occurrence", use greplm search --exhaustive (no ranking/limit/per-file cap); drop to raw grep only for a quick one-off confirmation of an exact string.

Read the full file on GitHub · 68 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. 2d ago First seen · 68 lines · 42 tokens per session scan C e4e28281be0e

Subscribe to this mod's changes

greplm-search is an agent published in the GitHub repository KhaledSMQ/greplm (14 stars, last pushed 21d ago), licensed MIT. It adds 42 tokens to every session and 1,066 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). It is 97% identical to greplm-search, differing in 7 lines, and is treated as a copy.