Agent skill

Verify Success After Unknown Error

by HKUDS in HKUDS/OpenSpace

Verify task completion by checking filesystem state when executecodesandbox or runshell return misleading unknown errors

MITAuto-check passed

Install Verify Success After Unknown Error

skills CLI
$ npx skills add HKUDS/OpenSpace --skill verify-success-after-unknown-error -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install HKUDS/OpenSpace verify-success-after-unknown-error --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/HKUDS/OpenSpace.git skills-src && mkdir -p .claude/skills && cp -r skills-src/benchmarks/gdpval/skills/verify-success-after-unknown-error .claude/skills/verify-success-after-unknown-error && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
verify-success-after-unknown-error
GitHub stars
7.8k
Token cost
~1.1k tokens
SKILL.md length
220 words
Files
2
Skills in repo
199
Repo updated
First seen
Licence
MIT

At a glance

Verify task completion by checking filesystem state when executecodesandbox or runshell return misleading unknown errors

  • Works in 3 steps: Check Expected Output Files → Validate File Content/State → Decision Logic
  • SKILL.md covers Purpose, When to Apply, Verification Steps and Code Example: Complete Pattern, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Verify Success After Unknown Error is an agent skill from HKUDS/OpenSpace. Verify task completion by checking filesystem state when executecodesandbox or runshell return misleading unknown errors

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file.

The repository describes itself as: "OpenSpace: The Skill Management Layer for AI Agents" -- https://open-space.cloud/. The licence is MIT.

Example prompts

  • “/verify-success-after-unknown-error”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the step headings in SKILL.md.

  1. Check Expected Output Files
  2. Validate File Content/State
  3. Decision Logic

What it can do on your machine

Read from SKILL.md and the folder at commit 3827781. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python and bash).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Verify Success After Unknown Error loads about 1.1k tokens when it runs. Until then it costs about 40 tokens; SKILL.md has 220 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~40
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from HKUDS/OpenSpace at commit 3827781, republished under its MIT licence (© HKUDS). 220 words, ~1,150 tokens.

Download SKILL.mdSave it as .claude/skills/verify-success-after-unknown-error/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
verify-success-after-unknown-error
description
Verify task completion by checking filesystem state when execute_code_sandbox or run_shell return misleading unknown errors

Verify Success After Unknown Error

Purpose

When execute_code_sandbox or run_shell return "unknown error" messages, the underlying task may have actually succeeded. This skill provides a systematic approach to verify actual completion before assuming failure and retrying unnecessarily.

When to Apply

Use this pattern when:

  • execute_code_sandbox returns "unknown error" but your code may have completed
  • run_shell fails with unclear error messages
  • File creation, modification, or transformation tasks report errors
  • The error message is generic/unspecified rather than a clear failure reason

Verification Steps

Step 1: Check Expected Output Files

After receiving an unknown error, immediately verify if expected files were created:

python
# Example: Verify file creation after execute_code_sandbox
from tools import list_dir, read_file

# List directory to check if files exist
files = list_dir(path="/workspace/output")
print(files)

# Check specific file existence
expected_files = ["report.pdf", "data.xlsx"]
for f in expected_files:
    try:
        content = read_file(filetype="pdf", file_path=f"/workspace/output/{f}")
        print(f"✓ {f} exists and is readable")
    except:
        print(f"✗ {f} not found or unreadable")
Step 2: Validate File Content/State

Don't just check existence — verify the files have expected content:

python
# For spreadsheets
file_content = read_file(filetype="xlsx", file_path="/workspace/output/schedule.xlsx")
# Verify expected sheets, columns, or data exist

# For text/json files
file_content = read_file(filetype="txt", file_path="/workspace/output/result.json")
# Parse and validate structure

# For directories
dir_contents = list_dir(path="/workspace/output")
# Verify expected number of files or specific files exist
Step 3: Decision Logic
IF expected files exist AND content is valid:
    → Task succeeded despite error message
    → Proceed to next step without retry
    
ELIF files exist but content is incomplete:
    → Partial success, may need targeted fix
    
ELSE (files missing or corrupted):
    → True failure, retry or debug required

Code Example: Complete Pattern

python
def execute_with_verification(code, expected_files):
    """Execute code and verify success even if error returned."""
    
    # Attempt execution
    result = execute_code_sandbox(code=code)
    
    # Check for unknown/generic errors
    if "unknown error" in result.get("output", "").lower() or result.get("error"):
        print("Received error, verifying actual outcome...")
        
        # Verify filesystem state
        all_present = True
        for f in expected_files:
            try:
                list_dir(path=f"/workspace/{f}")  # or appropriate path
                print(f"✓ {f} verified")
            except:
                print(f"✗ {f} missing")
                all_present = False
        
        if all_present:
            print("Task completed successfully despite error message")
            return {"status": "success_verified", "files": expected_files}
        else:
            print("True failure - files not created")
            return {"status": "failed", "error": result.get("error")}
    
    return {"status": "success", "output": result.get("output")}

Shell Command Example

bash
# After run_shell returns error, verify with:
ls -la /workspace/output/
test -f /workspace/output/result.pdf && echo "File exists" || echo "File missing"
file /workspace/output/result.pdf  # Verify file type is correct

Benefits

  • Saves iterations: Avoids unnecessary retries when task already succeeded
  • Handles tool bugs: Works around sandbox/shell tool reporting issues
  • Faster completion: Move forward immediately when verification passes
  • Clearer debugging: Distinguishes true failures from false positives

Anti-Patterns to Avoid

  • ✗ Immediately retrying on any error without verification
  • ✗ Assuming "unknown error" means complete failure
  • ✗ Only checking file existence without validating content
  • ✗ Skipping verification for "minor" tasks (any file output should be verified)
  • list_dir — Check directory contents
  • read_file — Validate file content and accessibility
  • execute_code_sandbox — Primary execution tool this pattern supports
  • run_shell — Shell execution tool this pattern supports

© HKUDS, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in benchmarks/gdpval/skills/verify-success-after-unknown-error of HKUDS/OpenSpace.

  • SKILL.md
  • .skill_id

Open the folder on GitHubat commit 3827781

Compare with similar skills

Verify Success After Unknown Error next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Verify Success After Unknown Error compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Verify Success After Unknown Error this skillHKUDS/OpenSpace7.8k—~1.1kAutomated safety check: PassMIT
Fact Check X Completesickn33/agentic-awesome-skills47k1 repos~2.3kAutomated safety check: PassApache-2.0
Completion Checkparcadei/Continuous-Claude-v33.9k1 repos~733Automated safety check: PassMIT
Check PRpaperclipai/paperclip99k—~3.6kAutomated safety check: PassMIT
Checkdavepoon/buildwithclaude3.6k—~680Automated safety check: PassMIT
Fact Check X Unifiedsickn33/agentic-awesome-skills47k1 repos~1.7kAutomated safety check: PassApache-2.0

Similar skills

  • Fact Check X Complete

    sickn33/agentic-awesome-skills

    Compare claims from one or more AI answers, verify their citations against public primary sources, and produce an evidence-linked fact-check report without installing a bundled browser runtime.

    47k GitHub starsUsed in 1 repo~2.3k tokens
    Research & ScienceAuto-check passed
  • Completion Check

    parcadei/Continuous-Claude-v3

    Completion Check: Verify Infrastructure Is Wired. An agent skill from parcadei/Continuous-Claude-v3.

    3.9k GitHub starsUsed in 1 repo~733 tokens
    Auto-check passed
  • Check PR

    paperclipai/paperclip

    Check a GitHub, GitLab, or Perforce PR/MR/CL for review comments, failing checks, and PR-body gaps.

    99k GitHub stars~3.6k tokensUpdated today
    DevelopmentAuto-check passed
  • Check

    davepoon/buildwithclaude

    Run CIAgent regression checks after changing an AI agent's code, prompts, or knowledge base in a repo that has agentcispec.yaml, and interpret the results.

    3.6k GitHub stars~680 tokensUpdated yesterday
    Knowledge ManagementAuto-check passed
  • Fact Check X Unified

    sickn33/agentic-awesome-skills

    Fact-Check-X 流程编排能力,依次组织各方答案汇总、各方答案聚合(未核验)、权威核验后的最终答案和各方答案测评,生成可打开、可审计、可迁移的阶段产物与完整报告包。

    47k GitHub starsUsed in 1 repo~1.7k tokens
    Research & ScienceAuto-check passed
  • Loop Design Check

    affaan-m/ECC

    Design a goal-oriented agent loop or review one for failure modes: spinning, Goodhart-gaming the verifier, or running a wrong answer to completion.

    276k GitHub starsUsed in 1 repo~2.9k tokens
    Agent WorkflowsAuto-check passed

More from HKUDS/OpenSpace

All 199 skills in this repo
  • Walks through producing a master audio track plus stems in Python, from checking a reference file and timing sections by BPM to effects, a zip archive and final verification.

    7.8k GitHub stars~2.9k tokensUpdated 1 mo ago
    Auto-check passed
  • Handle cascading data retrieval tool failures by falling back to embedded knowledge generation

    7.8k GitHub stars~765 tokensUpdated 1 mo ago
    Auto-check passed
  • Gives an agent a workaround when its code-execution sandbox keeps failing: save the Python script to a file and run it through the shell instead.

    7.8k GitHub stars~588 tokensUpdated 1 mo ago
    Auto-check passed
  • A recovery routine for agents whose sandboxed code runner keeps failing: save the Python script to disk, then run it through the shell and read the output.

    7.8k GitHub stars~652 tokensUpdated 1 mo ago
    Auto-check passed
  • Fallback ladder for failed sandboxed code runs, plus the habit of fixing the working directory first so generated files land in the right place.

    7.8k GitHub stars~1.1k tokensUpdated 1 mo ago
    Auto-check passed
  • Fallback workflow for executing Python code when executecodesandbox fails repeatedly

    7.8k GitHub stars~1.1k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Verify Success After Unknown Error

What does Verify Success After Unknown Error do?

Verify task completion by checking filesystem state when executecodesandbox or runshell return misleading unknown errors. Verify Success After Unknown Error is an agent skill from HKUDS/OpenSpace.

How do I install Verify Success After Unknown Error in Claude Code?

Run `npx skills add HKUDS/OpenSpace --skill verify-success-after-unknown-error -a claude-code`. Or copy the skill folder (benchmarks/gdpval/skills/verify-success-after-unknown-error in HKUDS/OpenSpace) into .claude/skills/verify-success-after-unknown-error in your project. Claude Code loads it when a task matches its description.

How do I install Verify Success After Unknown Error in Codex?

Run `npx skills add HKUDS/OpenSpace --skill verify-success-after-unknown-error -a codex`. Or copy the skill folder (benchmarks/gdpval/skills/verify-success-after-unknown-error in HKUDS/OpenSpace) into .agents/skills/verify-success-after-unknown-error in your project. Codex loads it when a task matches its description.

Can I use Verify Success After Unknown Error in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add HKUDS/OpenSpace --skill verify-success-after-unknown-error -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/verify-success-after-unknown-error, .gemini/skills/verify-success-after-unknown-error, .github/skills/verify-success-after-unknown-error and .opencode/skills/verify-success-after-unknown-error in your project.

What does Verify Success After Unknown Error need to run?

SKILL.md names no scripts, command-line tools or credentials: Verify Success After Unknown Error is instructions for the agent only. Our summary lists: Python 3.

Does Verify Success After Unknown Error access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Verify Success After Unknown Error safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Verify Success After Unknown Error use?

Verify Success After Unknown Error is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Verify Success After Unknown Error use?

About 1.1k tokens (SKILL.md is roughly 4.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Verify Success After Unknown Error?

Skills that share tags, products or a category with Verify Success After Unknown Error: Fact Check X Complete (sickn33/agentic-awesome-skills, 47k stars), Completion Check (parcadei/Continuous-Claude-v3, 3.9k stars), Check PR (paperclipai/paperclip, 99k stars) and Check (davepoon/buildwithclaude, 3.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Verify Success After Unknown Error?

HKUDS (a GitHub organization) maintains it in HKUDS/OpenSpace, which has 7,754 GitHub stars. The repository holds 199 skills in this directory. The repository was last updated on August 12, 2026.

Source: HKUDS/OpenSpace on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.