More screenshots
Works with the AI Agents you already use
Temporal Logic & Causal Auditor
A rigorous auditing skill that reconstructs incident timelines and subjects causal claims to formal logical tests.
$10
Temporal Logic & Causal Auditor
Example session with this skill installed
Analyze these logs.
[10:01:05] Srv-A: Timeout on DB-1.
[10:00:58] DB-1: Max connections reached.
[10:01:10] Eng-1: DB-1 crashed because Srv-A hammered it.
- Read your context and instructions
- Compiled the temporal logic
The intake detected 3 events across 2 systems (Srv-A, DB-1) spanning 12 seconds. Clock skew is not suspected.
Intake Check and Topology
- Provided: 3 events, 2 sources.
- Timestamp quality: No timezone or date provided. Range computed as UTC-12 to UTC+14.
- Topology: Srv-A depends on DB-1 (Observed in E1).
Timeline
| Ref | Timestamp (as stated) | Original timestamp | Source / System | Event description | Status |
|---|---|---|---|---|---|
| E1 | 10:00:58 | 10:00:58 | DB-1 | Max connections reached. | Observed |
| E2 | 10:01:05 | 10:01:05 | Srv-A | Timeout on DB-1. | Observed |
| E3 | 10:01:10 | 10:01:10 | Eng-1 | DB-1 crashed because Srv-A hammered it. | Observed |
Items in input: 3 = 3 rows + 0 undated + 0 statements + 0 excluded
Facts vs. Hypotheses
Verified Facts
- DB-1 reported max connections before Srv-A reported a timeout (E1, E2).
- Eng-1 made a causal claim at 10:01:10 (E3).
Causal Hypotheses
H1: Srv-A traffic caused DB-1 connection limit. (Low confidence: Temporal precedence fails; DB-1 hit limit before Srv-A timeout).
- H2: DB-1 exhaustion caused Srv-A timeouts. (High confidence: Supported by E1 and E2).
Contradictions
- Eng-1 claims Srv-A caused the crash (E3), but DB-1 was already at max capacity (E1) before Srv-A logged its error (E2).
Logical Tests
| Hyp | Temporal | Spatial | Alt Explanations | Negative Evidence |
|---|---|---|---|---|
| H1 | Fail (E1 < E2) | Pass (E1) | Unknown | E1 occurred before E2 |
| H2 | Pass | Pass | None found | None found |
Root Cause Assessment
The most supported root cause is DB-1 reaching its connection limit (E1), leading to Srv-A timeouts (E2). The claim in E3 is refuted by the timeline.
Risks and Action Plan
- Immediate Mitigation: Check DB-1 current connection count.
grep "Max connections" DB-1.log - Data Request: Review Srv-A request logs for 09:55:00 to 10:00:58 to see if traffic spiked before E1.
Next steps
- Provide logs from Srv-A prior to 10:00:00 to verify H1 traffic levels.
- Check DB-1 configuration for max_connections settings.
Connects securely to your tools. The creator never sees your data.
About this skill
Unstructured incident data often leads to "hallucinated" causality and correlation errors. This skill provides a rigorous framework for systems analysts to reconstruct timelines and audit the logic of incident reports. It forces a separation between observed facts, human claims, and causal hypotheses using a formal five-point logical test.
What it does
- Timeline reconstruction creates a normalized chronological table from mixed-source logs and messages.
- Timestamp auditing detects clock skew, timezone ambiguities, and missing date formats without silent assumptions.
- Causal verification subjects every hypothesis to tests for temporal precedence, system proximity, and negative evidence.
- Topology mapping extracts explicit service dependencies and communication paths from input text.
- Risk prediction identifies future failure points based on evidence-backed trends rather than generic advice.
How it works
- Intake and Topology: The agent scans logs to determine event counts, time spans, and stated system connections.
- Phase 1 Reconstruction: All events are mapped to a UTC-normalized table with strict status labels for observed vs. inferred data.
- Phase 2 Matrix: A fact-versus-hypothesis matrix separates verified evidence from engineer claims and causal theories.
- Phase 3 Logical Testing: Every surviving hypothesis is graded against five specific logical constraints to eliminate false correlations.
Frameworks & tools
This tool works with raw text, log files, and incident reports. It generates Markdown tables and can output grep, LogQL, or SQL queries for mitigation based on the detected stack.
Why this beats prompting it yourself
Standard LLM prompting often conflates "Event A happened before B" with "A caused B." This skill enforces a "Root Cause Assessment" protocol that discards hypotheses failing temporal or spatial tests, ensuring findings are traceable to specific log lines.
Use cases
- Auditing chaotic Slack transcripts and CloudWatch logs during a post-mortem.
- Reconciling mixed-timezone logs from global service clusters.
- Verifying whether a specific deploy actually preceded a reported spike in errors.
- Identifying gaps in telemetry where critical data for a root cause is missing.
Known limitations
Does not perform automated log ingestion via API; requires text-based logs or file access via tools. Cannot resolve timestamps if no timezone or offset is provided in the source.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.
Requires manual text input or file read. Cannot resolve absolute time for dateless logs. Does not support automated API-based log fetching.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 19 days ago