The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill mentions the use of METICULOUS_API_TOKEN and the configuration file ~/.meticulous/config.json for authentication. These are documented as standard methods for authenticating with the Meticulous platform and are not used in an unsafe manner.
[COMMAND_EXECUTION]: The skill instructs the user to run various meticulous CLI commands (e.g., meticulous local relevant-sessions, meticulous debug replay) and standard git commands (git add, git commit). These are part of the intended development workflow for the tool.
[INDIRECT_PROMPT_INJECTION]: The skill instructions involve processing session IDs and commit SHAs derived from the user's local development environment. While this represents a data ingestion surface, it is limited to specific CLI arguments for a developer tool and does not present a high risk of malicious injection in this context.

meticulous-iterative-dev