← Index

OpenGrok MCP Server

io.github.IcyHot09/opengrok-mcp-server·v9.0.1·Search

MCP server bridging OpenGrok search engine with AI for instant context across massive codebases.

Trust verdict · v1 advisory · method
NOT YET SCREENEDno verdict on file

Verdict not yet evaluated for this tool. The semantic screen takes adversarial cases first; coverage rolls out as the corpus expands (15/150 labels to graduation). The deterministic conformance probe is built but has not yet run on the public corpus, so a recorded verdict here is REVIEW or UNVERIFIED, never a clearing ALLOW. Until a verdict is recorded, an agent should treat this tool as not-yet-cleared and fall back to its own checks. Method: the eval, four-state verdict, honest limits.

Own this server? Screen its description →

Environment variables
OPENGROK_BASE_URL
required

OpenGrok server base URL (e.g. https://opengrok.example.com/source/)

OPENGROK_USERNAME
required

OpenGrok authentication username

OPENGROK_PASSWORD
secret

OpenGrok authentication password. Prefer storing via OS keychain using 'npx opengrok-mcp-server setup'.

OPENGROK_VERIFY_SSL

Verify TLS certificates. Set to false for self-signed or internal CA certs.

OPENGROK_CODE_MODE

Enable Code Mode: 5-tool sandbox interface with env.opengrok.* JavaScript API. Saves 80-95% tokens on large codebases.

OPENGROK_DEFAULT_PROJECT

Default project to scope all searches. Can be overridden per call.

OPENGROK_CONTEXT_BUDGET

Response size tier controlling token usage.

OPENGROK_ENABLE_ELICITATION

Enable interactive project picker at session start and env.opengrok.elicit() in Code Mode sandbox. Requires a supporting MCP client.

OPENGROK_SAMPLING_MODEL

Model preference for MCP Sampling (used for error explanation and query reformulation).

MCP quality score · maturity, not trust · methodology
freshness
22
completeness
20
installability
25
documentation
15
stability
10
Alternatives in Search