SKILLEMALL.ai
GitHub ↗ RU Sign in

CC agent-self-rollback

Deploy a self-rollback / snapshot mechanism for an AI agent's own memory files (identity SOUL.md, user profile USER.md, FACT.md, JOURNAL) and project files, so the agent can survive its own mistakes. Covers snapshotting, listing, restoring, and MD5 verify, plus self-discipline rules to write into the agent's own rules. Has a portable PowerShell script (snapshot/list/restore/verify). Use when the user asks for "agent rollback" "误操作回滚" "self-rollback mechanism" "protect my agent memory" "备份agent记忆" "快照回滚" or wants to harden an AI assistant against accidentally corrupting its own persistent memory. 关键词：误操作回滚、记忆保护、快照、self-rollback、agent memory backup、自毁预防。

ClawHub Agent Skills v1.0.0 5 files · 1 script body ≈ 913 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

Deploy a self-rollback / snapshot mechanism for an AI agent's own memory files (identity SOUL.md, user profile USER.md, FACT.md, JOURNAL) and project files…

As a process C 54/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice

AnalyzerAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
C
74/100
safety, quality, tests
Safety 60%
61
Quality 40%
93
Run on models
none yet
Process rating
C
54/100
Has gaps
Result and completion w 14
0
Inputs and preconditions w 11
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Dangerous commands medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.

For the author

Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.

How to improve

    For the model run — optional
    • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
    • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

    Guard findings · 11

    ✓ No critical or high findings

    Medium and low: 11
    • medium Dangerous commands cmd-execpolicy-bypass README.md:39
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 snapshot
    • medium Dangerous commands cmd-execpolicy-bypass README.md:42
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 snapshot
    • medium Dangerous commands cmd-execpolicy-bypass README.md:45
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 verify
    • medium Dangerous commands cmd-execpolicy-bypass README.md:48
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 restore 2
    • medium Dangerous commands cmd-execpolicy-bypass README.md:51
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 list
    • medium Dangerous commands cmd-execpolicy-bypass SKILL.md:52
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 snapshot
    • medium Dangerous commands cmd-execpolicy-bypass SKILL.md:73
      Runs PowerShell with execution policy bypassed
      powershell -NoProfile -ExecutionPolicy Bypass -File "%~dp0r…ps1" snapshot
    • low Dangerous commands cmd-execpolicy-bypass scripts/rollback.ps1:11
      Runs PowerShell with execution policy bypassed (code comment)
      #   powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 snapshot
      comment
    • low Dangerous commands cmd-execpolicy-bypass scripts/rollback.ps1:12
      Runs PowerShell with execution policy bypassed (code comment)
      #   powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 list
      comment
    • low Dangerous commands cmd-execpolicy-bypass scripts/rollback.ps1:13
      Runs PowerShell with execution policy bypassed (code comment)
      #   powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 restore
      comment
    • low Dangerous commands cmd-execpolicy-bypass scripts/rollback.ps1:14
      Runs PowerShell with execution policy bypassed (code comment)
      #   powershell -NoProfile -ExecutionPolicy Bypass -File roll…ps1 verify
      comment

    Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.

    Against the Agent Skills spec

    ✓ No remarks against the Agent Skills spec

    Process rating: all ten parameters 54/100

    • 0Result and completion. Does not say what the result is
    • 0Inputs and preconditions. Does not say what the process needs to start
    • 0Progress reporting. Says nothing while it works
    • 30Running it twice. 1 mutating operations with no state check
    • 50Failures and branches. 0 branches, has a failure section
    • 60Tools and files. Uses tools (bash) that frontmatter does not declare
    • 70When it triggers. States when to use, but not when not to
    • 100Steps. 35 steps
    • 100Consistency. Name and required fields are in place
    • 100Execution cost. Instruction body is 913 tokens

    Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

    Quality signals

    • +4Description does not say when NOT to use the skill (false activations)
    • +3Output format is not stated: the model decides each time
    • +2Single-language instructions
    • +5Description quotes 4 example trigger phrases
    • +3Description length 660: enough signal without eating the budget
    • +4Structure: 15 headings
    • +3Step-by-step instructions: 35 items
    • +4Has examples (2 code blocks)
    • +3All 1 scripts are documented
    • +1License stated

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 93.

    External checks

    ClawHub: suspicious
    This rollback skill is coherent and not malicious, but it handles sensitive agent memory and project files with broad local copying and weak snapshot containment that users should review carefully.
    LLM: suspicious (high) · VirusTotal: