The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute several shell-based tools including adlc-cli, run-branch.sh, and open-mr.sh to manage workflow states and repository operations. These commands are integral to the skill's operation within its defined ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading external documents (spec.md and plan.md) to drive the generation of tests and source code, creating a surface for injection attacks.
Ingestion points: Input is ingested from spec.md and plan.md during the test and code phases (identified in SKILL.md).
Boundary markers: The instructions lack explicit requirements for the agent to use delimiters or to ignore instructions embedded within the processed specification files.
Capability inventory: The skill possesses capabilities to write to the local filesystem (src/ and tests/), execute shell scripts, and interact with the adlc-cli workflow manager.
Sanitization: There is no mention of sanitizing or validating the contents of the external markdown files before they are interpreted to generate code.

mission-implement