wayback-link-fallback
Skill1 starsChanged 5 months agoMITView on GitHub
Use when a task depends on a dead, redirected, incomplete, version-specific, or historically changed web page and archived evidence from the Wayback Machine may recover the required source.
In this repository14 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-skills-archive-orchestratorSkill
- call-r-lib-in-pythonSkill
- code-commenterSkill
- code-reviewerSkill
- end-user-xp-improverSkill
- fp-developerSkill
- lesson-documenterSkill
- plan-designerSkill
- preferred-workflowSkill
- release-preparerSkill
- scientific-prose-editorSkill
- simple-code-writerSkill
- spec-driven-developerSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
If historical or unavailable content is needed, query captures for the exact original URL.
- Distinctive: line 22
Verify current claims against a live primary source.
- Distinctive: line 34
Avoid broad wildcard crawls or asset enumeration unless explicitly required.
- Distinctive: line 38
If no reliable capture exists, try one narrower primary alternative: versioned official docs, a tagged source file, package registry, …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Wayback Link Fallback
Recover the smallest reliable archived source needed for the task and label it as historical evidence.
Workflow
- Try the original URL and determine whether the live content still answers the question.
- If historical or unavailable content is needed, query captures for the exact original URL. Prefer one narrow CDX query over manually opening many snapshots.
- Select a successful capture that matches the relevant date, product version, and MIME type; avoid login pages, soft 404s, consent screens, and unrelated redirects.
- Fetch raw archived content with the
id_form when possible to avoid archive UI and URL rewriting. - Record the original URL, snapshot URL, capture timestamp, and any version or completeness uncertainty.
- Verify current claims against a live primary source. Never present an archive snapshot as current truth merely because it was retrievable.
For CDX parameters, capture selection details, and relative-link resolution, read capture and link handling.
Request discipline
- Make the fewest requests that can answer the task.
- Collapse duplicate digests in capture-heavy results.
- Back off on slow, empty,
429, or503responses. - Cache useful timestamps and URLs in task-local notes instead of repeating requests.
- Avoid broad wildcard crawls or asset enumeration unless explicitly required.
Failure handling
If no reliable capture exists, try one narrower primary alternative: versioned official docs, a tagged source file, package registry, release page, official mirror, or repository history. Keep fallback provenance explicit.
Report failure rather than relying on captures that are blocked, partial, mismatched to the target version, or missing required scripts, assets, schemas, or linked pages.
For legal, medical, financial, security-sensitive, or other high-stakes current claims, use archived content only as historical evidence and verify the present state with live authoritative sources.
Expected output
Include the original URL, archive URL, capture timestamp (YYYYMMDDhhmmss), relevant human
date, why that capture was chosen, and all material uncertainty.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

