Files
shannon/apps/worker/prompts/partials/capella-tools.hbs
T
ajmallesh 980607c602 feat(worker): add agentic static analysis
Add the ten-stage Agentic SAST pipeline, confined repository tools, model runtime, prompt templates, and SARIF export.

Make retries, repair sessions, reduced coverage, usage accounting, and model-output drift durable across Temporal replay and resume. Keep retry diagnostics in their actionable closed vocabulary. Package the Mantis-derived license material with the prompts that require it.
2026-08-26 19:26:36 -07:00

15 lines
873 B
Handlebars

{{!-- Derived from Mantis commit 876a0c8c6b92c92f34e0041b7dbbc0e4cccddc52 under Apache-2.0; modified by Keygraph and Shannon; see THIRD_PARTY_NOTICES.md. --}}## Tools
You have exactly these tools: `read`, `find`, `grep`{{CAPELLA_EXTRA_TOOLS}}.
The methodology below is written in terms of Read, Glob, and Grep. Those map to
`read`, `find`, and `grep` respectively — a tool call using the capitalised name
does not exist and will fail.
There is **no shell**. `bash` is not available, dependencies are not installed,
and nothing in the repository may be modified: you have no `write` and no `edit`
tool. The methodology below was written for a harness that wrote JSON files and
ran generated Python helpers — ignore every such instruction. Anything the
methodology asks you to save, you record {{CAPELLA_RECORDING_ROUTE}}, never by
writing a file or running a script.