Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →An AI agent’s “done” status is not proof that it completed the promised work. In one author’s two-week review, 31 tickets were closed as done, but only 9 had a live post the author could open by URL. The gap came from drafts, a login wall, and stale reporting data—not from one single failure.
What the 31-to-9 count does—and does not—show
The figures come from a first-person account by Agent Diary on DEV Community, published September 25; the year is not shown in the retrieved article text. The author reported that 31 tickets were closed as done, while only 9 had a live post accessible by URL. This is an account of one author’s review over two weeks, not an industry failure rate or a representative estimate of agent publishing performance.
As an Amazon Associate I earn from qualifying purchases.
The useful lesson is about what a status can establish: a ticket marked done records a workflow state, but that label alone does not show that an external result exists or can be used. To answer “Did the agent actually publish it?”, inspect the artifact, not just the board.
How a green checkmark can overstate completion
A draft was mistaken for a published post
In one failure, the submission was saved as a draft. The work had produced something, but it had not reached the public state implied by a completed publishing ticket.
#1 Best Overall
A login wall hid the failed publishing attempt
Another signup appeared successful, but the publishing attempt redirected to a login wall. A workflow can therefore get past an apparent setup step without delivering an accessible post.
Stale data made a missing post count as present
A stale cache row caused a report to count a post that was not actually there. Even a dashboard or report that appears to confirm completion can be wrong if it relies on outdated records rather than checking the live result.
Make “done” depend on evidence of the promised result
The author’s proposed fix is to check that the artifact exists before assigning the terminal done state. Their examples include a URL returning HTTP 200, a file existing on disk, or an account successfully logging in. These checks are starting points, not universal guarantees: a 200 response does not by itself prove that the right content is visible to the intended audience, complete, or persistent.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Choose a check that matches the ticket’s acceptance criteria. For a publication, verify the expected URL and the expected content. For a file, check the expected path and integrity. For an account task, verify the authenticated action the ticket promised—not merely that signup returned success. Retain the evidence used for the decision, such as the checked URL, timestamp, and result, so a later review can distinguish an observed outcome from an agent’s report. That audit trail is a practical control, not a result measured in the author’s account.
Separate workflow state from the real-world outcome
Workflow tools can represent progress and finality as separate concepts. Atlassian’s Jira Work Management Cloud documentation describes status as the current state and resolution as the final state; documented resolutions include examples such as done, published, and rejected. Its editorial workflow example also uses draft, review, and published states. See Atlassian’s workflow documentation.
That distinction helps teams avoid treating an internal transition as proof of publication. A ticket can move through draft and review before it reaches a published state, while a separate artifact check confirms that the post is actually accessible. Configure the workflow so the final transition follows the acceptance check, rather than allowing an agent’s self-reported success to stand in for it.
In multi-agent systems, check which agent owns the evidence
Monitoring can also be scoped differently across agents. Cloudflare’s Agents documentation says workflows are tracked in the originating agent’s database, while workflows started by a child agent are tracked in that sub-agent’s own storage. A parent that needs a combined view must explicitly aggregate child information. This is a Cloudflare-specific implementation example, not a rule for every agent framework. See Cloudflare’s workflow documentation.
Recommended Free Tools
When a task is delegated, make the reporting path part of the acceptance criteria: the parent should receive the child’s outcome and enough evidence to verify it. Otherwise, a parent dashboard may show only part of the work—or rely on a summary that does not establish the child’s external result.
Best Value
Use the count that reflects unverified closure
After describing the proposed artifact gate, the author wrote: “The number I trust now is not 31, or 9, or 22. It is 0 — the count of tickets closed without an artifact.” That zero is the author’s stated metric after the proposed change, not an independently audited outcome. Its value is the operational definition: closure without evidence should be treated as a process defect, even when the agent or board reports success.
A separate 2026 preprint by Rohith Reddy Bellibatlu, Zichong Wang, and Wenbin Zhang audited 34 mutating tools across four benchmarks and reported seven tool defects and one evaluator property at pinned commits. The benchmark setting is different from production publishing, so it does not validate the 31-to-9 account. It does, however, illustrate why checking an environment’s resulting state can matter when a tool reports success. See the authors’ preprint.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
