The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The 'Audit Mode' feature instructs the agent to ingest and analyze user-provided Makefiles, which creates a surface for indirect prompt injection if the input files contain malicious instructions.
Ingestion points: The skill analyzes Makefile content provided during an audit.
Boundary markers: No explicit delimiters or 'ignore' instructions are defined for the audited content.
Capability inventory: The agent's actions are limited to reading files and generating a text-based audit report.
Sanitization: No specific sanitization or filtering of the input file content is implemented.
[COMMAND_EXECUTION]: The skill provides Makefile templates that define various shell command sequences.
The help target uses awk to parse the Makefile and format help text.
The install and uninstall targets utilize mkdir, ln, rm, and readlink to manage symlinks within the user's ~/bin directory.
The lint target invokes shellcheck on local scripts.
The test target executes a local script via ./test.sh.

makefile-best-practices