self-managed-context

Pass

Audited by Gen Agent Trust Hub on Oct 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a self-managed context protocol where the agent has write access to a mirror of its own conversation history in a local file. This mechanism creates a surface for instructions to be persisted and potentially followed by the model in subsequent turns.
  • Ingestion points: The mirror file, located at /tmp/.live_ctx/LIVE_CTX_MAIN.txt, is read and parsed back into the message list during the execution step described in SKILL.md and references/harness-protocol.md.
  • Boundary markers: The implementation uses structured turn headers ([[CTX_TURN n role=...]]) and a "pinned prefix" system to ensure that core system instructions and the initial task are never exposed to the editable file or overwritten.
  • Capability inventory: The agent utilizes shell commands (run_shell) and file system operations to read, write, and manipulate the context mirror.
  • Sanitization: The parse_back logic includes "role folding," which automatically maps any model-written roles (including 'system' or 'tool') back to the 'user' role, preventing the agent from programmatically assuming higher authority.
  • [COMMAND_EXECUTION]: The skill provides functional examples where the agent uses shell commands to interact with its environment to perform context housekeeping.
  • Evidence: The step() function in SKILL.md executes commands provided by the model, and Example 2 demonstrates the use of shell-based text processing (via re.sub) to update the context file.
  • [EXTERNAL_DOWNLOADS]: The documentation references external research and code repositories from well-known and reputable organizations.
  • Evidence: Links and references to github.com/facebookresearch/context-language-models, arxiv.org, and safety reports from alignment.openai.com are used for technical validation.
  • [DYNAMIC_EXECUTION]: The reference implementation involves generating and running short scripts to perform surgical edits on the context mirror.
  • Evidence: Example 2 in SKILL.md shows a pattern where a Python script is generated as a heredoc within a Bash command to modify the mirror file at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 3, 2026, 08:23 PM