Paper Citation Record · LEDGER

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents

As of 20 August 2026, this Paper Citation Record lists 36 of 36 outbound references and 0 inbound Pith citation observations for arXiv:2608.11888.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2608.11888 v1

Coverage vector

measured 36 of 36 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T00:26:36.964619Z

measured 36 of 36 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-19T06:32:44.657259+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

36 of 36 outbound references displayed

verified exact0
verified fuzzy22
unresolved14
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 880a41a7-b057-477a-a641-ac2b56d97433 · outbound

This paper cites SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.830531Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3ccb5c91-4931-405b-acf9-b5167a0c7fba · outbound

This paper cites Swe-skills- bench: Do agent skills actually help in real-world software engineering?.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Swe-skills- bench: Do agent skills actually help in real-world software engineering?

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.835083Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d304092-63ab-4396-911d-6e9fbd363b04 · outbound

This paper cites Lost in the middle: How language models use long contexts,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Lost in the middle: How language models use long contexts,

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.501567Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 6b2d0fdb-26a6-4c9d-aaa6-7252deb63e06 · outbound

This paper cites Large language models can be easily distracted by irrelevant context,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Large language models can be easily distracted by irrelevant context,

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.843224Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc9253a3-202f-41f4-8214-72a7ac68772c · outbound

This paper cites Quantifying language models’ sensitivity to spurious features in prompt design or: How i learned to start worrying about prompt formatting,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Quantifying language models’ sensitivity to spurious features in prompt design or: How i learned to start worrying about prompt formatting,

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.480516Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 39a5e884-301e-48c9-896c-acd4d461b098 · outbound

This paper cites More doc- uments, same length: Isolating the challenge of multiple documents in rag,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents More doc- uments, same length: Isolating the challenge of multiple documents in rag,

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.850899Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5fe4a77f-ffcf-4967-b0da-39d15fdacaaf · outbound

This paper cites Differential testing for software,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Differential testing for software,

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.461786Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 899f2ebc-47de-4604-a7d0-98e4f6010311 · outbound

This paper cites Equipping agents for the real world with agent skills,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Equipping agents for the real world with agent skills,

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.448393Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 35bca5fa-fd82-448d-892c-47827473563d · outbound

This paper cites OpenCode: The open source coding agent,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents OpenCode: The open source coding agent,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.437465Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 6346949e-831b-4403-a265-b1612ef26aae · outbound

This paper cites Introducing claude opus 4.6,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Introducing claude opus 4.6,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.426617Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation ef7072a1-38dc-4a28-a1c0-7d7d76fe5171 · outbound

This paper cites Introducing gpt-5.5,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Introducing gpt-5.5,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.416047Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation c8db957b-9181-465c-8170-6f4d4bba60b7 · outbound

This paper cites Ensemble methods in machine learning,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Ensemble methods in machine learning,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.405126Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 804f795f-296f-4e4b-a27e-dba9b01f30f7 · outbound

This paper cites Self-Consistency Improves Chain of Thought Reasoning in Language Models.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Self-Consistency Improves Chain of Thought Reasoning in Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.879915Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c92d4d58-f465-432e-9e65-a6430f05a91f · outbound

This paper cites Voyager: An Open-Ended Embodied Agent with Large Language Models.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Voyager: An Open-Ended Embodied Agent with Large Language Models

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.884222Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b92d28f2-2adb-47fb-8579-e06b17052d83 · outbound

This paper cites Executable code actions elicit better llm agents,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Executable code actions elicit better llm agents,

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.888921Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8fbc1d13-ac23-42e8-8b33-a4929bb13464 · outbound

This paper cites Gpt4tools: Teaching large language model to use tools via self-instruction,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Gpt4tools: Teaching large language model to use tools via self-instruction,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.386875Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 92d17c18-a0c1-4eec-aa6a-5f2a880bc79b · outbound

This paper cites ReAct: Synergizing Reasoning and Acting in Language Models.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents ReAct: Synergizing Reasoning and Acting in Language Models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.896497Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1310748-b14b-4210-ac1d-08f76de121ce · outbound

This paper cites Same task, more tokens: the impact of input length on the reasoning performance of large language models,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Same task, more tokens: the impact of input length on the reasoning performance of large language models,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.374726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 03e8beaa-7204-4700-96e7-570a8d7f02f3 · outbound

This paper cites Why do multi-agent llm systems fail?.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Why do multi-agent llm systems fail?

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.904068Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf1d2fc8-ca0a-41b7-b68d-53691326170b · outbound

This paper cites Agentless: Demystifying LLM-based Software Engineering Agents.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Agentless: Demystifying LLM-based Software Engineering Agents

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.907676Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6463b77a-1cf8-4f4a-b391-c5dfc1859dd0 · outbound

This paper cites Swe-agent: Agent-computer interfaces enable automated software engineering,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Swe-agent: Agent-computer interfaces enable automated software engineering,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.356205Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 5bc29c56-70f1-48ab-84e1-f501c9cfda0b · outbound

This paper cites Swe-bench: Can language models resolve real-world github issues?.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Swe-bench: Can language models resolve real-world github issues?

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.344590Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation ab050f45-5cea-44e7-af9b-674eae1fa51f · outbound

This paper cites OpenHands: An Open Platform for AI Software Developers as Generalist Agents.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents OpenHands: An Open Platform for AI Software Developers as Generalist Agents

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.919199Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e805c5d-84d0-4549-bce8-35fbe75cbd37 · outbound

This paper cites Multi-swe-bench: A multilingual benchmark for issue resolving,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Multi-swe-bench: A multilingual benchmark for issue resolving,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.332498Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 9166c087-402a-4c56-99f9-2ea25e418323 · outbound

This paper cites SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.926696Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9cb7da78-2257-48de-a27f-b6a68ab0ca69 · outbound

This paper cites Measuring ai ability to complete long software tasks,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Measuring ai ability to complete long software tasks,

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.317630Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation b96ea37b-c03d-469c-8da3-06c81b357a64 · outbound

This paper cites Understanding and detecting real-world performance bugs,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Understanding and detecting real-world performance bugs,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.305473Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation aab1f957-4ef4-497c-aa30-a1be108ea07f · outbound

This paper cites Discovering, reporting, and fixing perfor- mance bugs,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Discovering, reporting, and fixing perfor- mance bugs,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.294977Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 37fceef0-e62d-4b25-8ad1-cf135d4532d1 · outbound

This paper cites Performance issues and optimizations in javascript: an empirical study,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Performance issues and optimizations in javascript: an empirical study,

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.282486Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation d7045bb6-25f1-4f10-965b-4f946cdc590a · outbound

This paper cites Understanding performance problems in deep learning systems,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Understanding performance problems in deep learning systems,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.270364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation f2735e11-491d-42db-905b-cb677e01f4dd · outbound

This paper cites A comprehensive study on deep learning bug characteristics,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents A comprehensive study on deep learning bug characteristics,

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.253130Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 0c106420-0050-4149-ac60-09e30d9750cf · outbound

This paper cites Taxonomy of real faults in deep learning systems,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Taxonomy of real faults in deep learning systems,

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.951615Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cad5573d-0fcc-4644-bad8-cec289311a6a · outbound

This paper cites An empirical study on tensorflow program bugs,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents An empirical study on tensorflow program bugs,

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-16T00:26:36.955099Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b6132151-456e-4736-b5b3-6abaced6b2b2 · outbound

This paper cites An empirical study on configuration errors in commercial and open source systems,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents An empirical study on configuration errors in commercial and open source systems,

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.224208Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 8eb61ccd-7549-48f4-bcc6-eb991ce5e6df · outbound

This paper cites Hey, you have given me too many knobs!: Understanding and dealing with over-designed configuration in system software,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents Hey, you have given me too many knobs!: Understanding and dealing with over-designed configuration in system software,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.212836Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 3cb0f87a-665b-4040-a1ae-60f9b1a61530 · outbound

This paper cites An empirical study on performance bugs for highly config- urable software systems,.

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents An empirical study on performance bugs for highly config- urable software systems,

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:26:37.200380Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Pith citing papers

No inbound Pith citation observations are available.