Nm Leyline Damage Control
v1.0.0Recover from broken agent state including crash recovery, context overflow, merge conflicts, and corrupted worktrees
⭐ 0· 44·1 current·1 all-time
by@athola
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
high confidencePurpose & Capability
The name and description (recovering crashed agents, context overflow, merge conflicts, and corrupted worktrees) align with the instructions, which focus on inspecting the task list, git history/index, and on-disk artifacts. The declared config path (night-market.error-patterns) is plausible for error classification and relates to the stated purpose.
Instruction Scope
SKILL.md and module files explicitly instruct reading the task list, running git commands (log, diff, status, stash, restore, etc.), producing handoff summaries, and documenting actions. All recommended actions fall inside the stated recovery remit; there are no instructions to exfiltrate data, call remote endpoints, or access unrelated system resources. Users should note the skill assumes access to the repository and task-list storage.
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so nothing is written to disk by an installer. Low install risk.
Credentials
The skill declares no required environment variables or credentials. It does require a config path (night-market.error-patterns), which is reasonable for error-pattern classification. No unrelated secrets or external service credentials are requested.
Persistence & Privilege
always is false and autonomous invocation is allowed (platform default). The skill does not request persistent system changes or to modify other skills' configuration. Procedures involve updating task records and git commits/stashes, which are appropriate for the stated recovery operations.
Assessment
This skill is coherent and instruction-only, but before using it: ensure the agent has appropriate (least-privilege) access to the repository and to whatever task-list service you use; back up/clone the working tree before applying automated reconcilation steps; provide or review the referenced night-market.error-patterns config so error classification matches your environment; and review the original repository (claude-night-market/leyline) if you want implementation context. If you plan to allow the agent to invoke this skill autonomously, ensure workflow guards and human-escalation paths are in place (the skill itself recommends escalation for ambiguous conflicts).Like a lobster shell, security has layers — review code before you run it.
latestvk97bxkxm8z0qf6s62v2rp5ndg184sbq6
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🦞 Clawdis
Confignight-market.error-patterns